--- name: "open" section: "2" platform: "freebsd" package: "base" version: "16.0-CURRENT-20261027-61c70dc16a79" release: "FreeBSD 16.0-CURRENT (2026-10-27)" platforms: ["linux", "freebsd", "macos"] url: "https://new.carta.tech/open/2" --- # open(2) open or create a file for reading, writing or executing ## NAME **open**, **openat** – open or create a file for reading, writing or executing ## LIBRARY Standard C Library (libc, -lc) ## SYNOPSIS **#include \** *int*\ **open**(*const char \*path*, *int flags*, *...*); *int*\ **openat**(*int fd*, *const char \*path*, *int flags*, *...*); ## DESCRIPTION The file name specified by *path* is opened for either execution or reading and/or writing as specified by the argument *flags* and the file descriptor returned to the calling process. The *flags* argument may indicate the file is to be created if it does not exist (by specifying the O\_CREAT flag). In this case **open**() and **openat**() require an additional argument *mode\_t mode*, and the file is created with mode *mode* as described in [chmod(2)](/chmod/2.md) and modified by the process' umask value (see [umask(2)](/umask/2.md)). The **openat**() function is equivalent to the **open**() function except in the case where the *path* specifies a relative path or the *O\_EMPTY\_PATH* flag is specified. For **openat**() and relative *path*, when *fd* references a directory, and the *O\_EMPTY\_PATH* flag is not specified, the file to be opened is determined relative to the directory associated with the file descriptor *fd* instead of the current working directory. The *flag* parameter and the optional fourth parameter correspond exactly to the parameters of **open**(). If **openat**() is passed the special value AT\_FDCWD in the *fd* parameter, the current working directory is used and the behavior is identical to a call to **open**(). When **openat**() is called with an absolute *path*, it ignores the *fd* argument. When **openat**() is called with the *fd* argument that does not reference a directory, the call fails unless *O\_EMPTY\_PATH* flag is specified, see below. In [capsicum(4)](/capsicum/4.md) capability mode, **open**() is not permitted. The *path* argument to **openat**() must be strictly relative to a file descriptor *fd*; that is, *path* must not be an absolute path and must not contain ".." components which cause the path resolution to escape the directory hierarchy starting at *fd*. Additionally, no symbolic link in *path* may target absolute path or contain escaping ".." components. *fd* must not be AT\_FDCWD. If the vfs.lookup\_cap\_dotdot [sysctl(3)](/sysctl/3.md) MIB is set to zero, ".." components in the paths, used in capability mode, are completely disabled. If the vfs.lookup\_cap\_dotdot\_nonlocal MIB is set to zero, ".." is not allowed if found on non-local filesystem. The *flags* are formed by *or*'ing the following values: - O\_RDONLY open for reading only - O\_WRONLY open for writing only - O\_RDWR open for reading and writing - O\_EXEC open for execute only - O\_SEARCH open for search only (an alias for O\_EXEC typically used with O\_DIRECTORY) - O\_NONBLOCK do not block on open - O\_APPEND set file pointer to the end of the file before each write - O\_CREAT create file if it does not exist - O\_TRUNC truncate size to 0 - O\_EXCL fail if O\_CREAT is set and the file exists - O\_SHLOCK atomically obtain a shared lock - O\_EXLOCK atomically obtain an exclusive lock - O\_DIRECT read and write directly from the backing store - O\_FSYNC synchronous data and metadata writes (historical synonym for O\_SYNC) - O\_SYNC synchronous data and metadata writes - O\_DSYNC synchronous data writes - O\_NOFOLLOW do not follow symlinks - O\_NOCTTY ignored - O\_TTY\_INIT ignored - O\_DIRECTORY error if file is not a directory - O\_CLOEXEC automatically close file on [execve(2)](/execve/2.md) - O\_CLOFORK automatically close file on any child process created with **fork**(*2*) - O\_VERIFY verify the contents of the file with mac\_veriexec(4) - O\_RESOLVE\_BENEATH ([openat(2)](/openat/2.md) only) path resolution must not cross the *fd* directory - O\_PATH record only the target path in the opened descriptor - O\_EMPTY\_PATH ([openat(2)](/openat/2.md) only) open file referenced by *fd* if path is empty - O\_NAMEDATTR open a named attribute or named attribute directory Exactly one of the flags O\_RDONLY, O\_WRONLY, O\_RDWR, or O\_EXEC must be provided. Opening a file with O\_APPEND set causes each write on the resulting file descriptor to be appended to the end of the file. If O\_TRUNC is specified and the file exists, the file is truncated to zero length. If O\_CREAT is set, but file already exists, this flag has no effect except when O\_EXCL is set too, in this case **open**() fails with EEXIST. This may be used to implement a simple exclusive access locking mechanism. In all other cases, the file is created and the access permission bits (see [chmod(2)](/chmod/2.md)) of the file mode are set to the value of the third argument taken as *mode\_t mode* and passed through the [umask(2)](/umask/2.md). This argument does not affect whether the file is opened for reading, writing, or for both. The open' request for a lock on the file, created with O\_CREAT, will never fail provided that the underlying file system supports locking; see also O\_SHLOCK and O\_EXLOCK below. If O\_EXCL is set and the last component of the pathname is a symbolic link, **open**() will fail even if the symbolic link points to a non-existent name. If O\_NONBLOCK is specified and the **open**() system call would block for some reason (for example, waiting for carrier on a dialup line), **open**() returns immediately. The descriptor remains in non-blocking mode for subsequent operations. If O\_SYNC is used in the mask, all writes will immediately and synchronously be written to disk. O\_FSYNC is an historical synonym for O\_SYNC. If O\_DSYNC is used in the mask, all data and metadata required to read the data will be synchronously written to disk, but changes to metadata such as file access and modification timestamps may be written later. If O\_NOFOLLOW is used in the mask and the target file passed to **open**() is a symbolic link then the **open**() will fail. When opening a file, a lock with [flock(2)](/flock/2.md) semantics can be obtained by setting O\_SHLOCK for a shared lock, or O\_EXLOCK for an exclusive lock. O\_DIRECT may be used to minimize or eliminate the cache effects of reading and writing. The system will attempt to avoid caching the data you read or write. If it cannot avoid caching the data, it will minimize the impact the data has on the cache. Use of this flag can drastically reduce performance if not used with care. The semantics of this flag are filesystem dependent, and some filesystems may ignore it entirely. O\_NOCTTY may be used to ensure the OS does not assign this file as the controlling terminal when it opens a tty device. This is the default on FreeBSD, but is present for POSIX compatibility. The **open**() system call will not assign controlling terminals on FreeBSD. O\_TTY\_INIT may be used to ensure the OS restores the terminal attributes when initially opening a TTY. This is the default on FreeBSD, but is present for POSIX compatibility. The initial call to **open**() on a TTY will always restore default terminal attributes on FreeBSD. O\_DIRECTORY may be used to ensure the resulting file descriptor refers to a directory. This flag can be used to prevent applications with elevated privileges from opening files which are even unsafe to open with O\_RDONLY, such as device nodes. O\_CLOEXEC may be used to set FD\_CLOEXEC flag for the newly returned file descriptor. O\_CLOFORK may be used to set FD\_CLOFORK flag for the newly returned file descriptor. The file will be closed on any child process created with **fork**(*2*), **vfork**(*2*) or **rfork**(*2*) with the RFFDG flag, remaining open in the parent. Both the O\_CLOEXEC and O\_CLOFORK flags can be modified with the F\_SETFD **fcntl**(*2*) command. O\_VERIFY may be used to indicate to the kernel that the contents of the file should be verified before allowing the open to proceed. The details of what “verified” means is implementation specific. The run-time linker (rtld) uses this flag to ensure shared objects have been verified before operating on them. O\_RESOLVE\_BENEATH returns ENOTCAPABLE if any intermediate component of the specified relative path does not reside in the directory hierarchy beneath the starting directory. Absolute paths or even the temporal escape from beneath of the starting directory is not allowed. When a directory is opened with O\_SEARCH, execute permissions are checked at open time. The returned file descriptor may not be used for any read operations like [getdirentries(2)](/getdirentries/2.md). The primary use of this descriptor is as the lookup descriptor for the **\*at**() family of functions. If O\_SEARCH was not requested at open time, then the **\*at**() functions use the current directory permissions for the directory referenced by the descriptor at the time of the **\*at**() call. O\_PATH returns a file descriptor that can be used as the first argument for **openat**() and other filesystem-related system calls collectively named **\*at**() taking a file descriptor argument, like [fstatat(2)](/fstatat/2.md) and others. The other functionality of the returned file descriptor is limited to the following descriptor-level operations: - [connectat(2)](/connectat/2.md) for unix domain socket (see [unix(4)](/unix/4.md)) - [fcntl(2)](/fcntl/2.md) but advisory locking is not allowed - [dup(2)](/dup/2.md) - [close(2)](/close/2.md) - [fstat(2)](/fstat/2.md) - [fstatfs(2)](/fstatfs/2.md) - [fchdir(2)](/fchdir/2.md) - [fchroot(2)](/fchroot/2.md) - [fexecve(2)](/fexecve/2.md) - [funlinkat(2)](/funlinkat/2.md) can be passed as the third argument - SCM\_RIGHTS can be passed over a [unix(4)](/unix/4.md) socket using a SCM\_RIGHTS message - [kqueue(2)](/kqueue/2.md) only with EVFILT\_VNODE - \_\_acl\_get\_fd(2) - \_\_acl\_aclcheck\_fd(2) - [extattr(2)](/extattr/2.md) - [capsicum(4)](/capsicum/4.md) can be passed to **cap\_\*\_limit**() and **cap\_\*\_get**() system calls (such as [cap\_rights\_limit(2)](/cap_rights_limit/2.md)). Other operations like [read(2)](/read/2.md), [ftruncate(2)](/ftruncate/2.md), and any other that operate on file and not on file descriptor (except [fstat(2)](/fstat/2.md)), are not allowed. A file descriptor created with the O\_PATH flag can be opened as a normal (operable) file descriptor by specifying it as the *fd* argument to **openat**() with an empty *path* and the O\_EMPTY\_PATH flag. Such an open behaves as if the current path of the file referenced by *fd* is passed, except that path walk permissions are not checked. See also the description of AT\_EMPTY\_PATH flag for [fstatat(2)](/fstatat/2.md) and related syscalls. Conversely, a file descriptor fd referencing a filesystem file can be converted to the O\_PATH type of descriptor by using the following call ``` opath_fd = openat(fd, "", O_EMPTY_PATH | O_PATH); ``` If successful, **open**() returns a non-negative integer, termed a file descriptor. It returns -1 on failure. The file descriptor value returned is the lowest numbered descriptor currently not in use by the process. The file pointer used to mark the current position within the file is set to the beginning of the file. If a sleeping open of a device node from [devfs(4)](/devfs/4.md) is interrupted by a signal, the call always fails with EINTR, even if the SA\_RESTART flag is set for the signal. A sleeping open of a fifo (see [mkfifo(2)](/mkfifo/2.md)) is restarted as normal. When a new file is created, it is assigned the group of the directory which contains it. Unless O\_CLOEXEC flag was specified, the new descriptor is set to remain open across [execve(2)](/execve/2.md) system calls; see [close(2)](/close/2.md), [fcntl(2)](/fcntl/2.md) and the description of the O\_CLOEXEC flag. When the O\_NAMEDATTR flag is specified for an **openat**() where the *fd* argument is for a file object, a named attribute for the file object is opened and not the file object itself. If the O\_CREAT flag has been specified as well, the named attribute will be created if it does not exist. When the O\_NAMEDATTR flag is specified for a **open**(), a named attribute for the current working directory is opened and not the current working directory. The *path* argument for this **openat**() or **open**() must be a single component name with no embedded ‘/’. If the *path* argument is ‘.’ then the named attribute directory for the file object is opened. (See [named\_attribute(7)](/named_attribute/7.md) for more information.) The system imposes a limit on the number of file descriptors open simultaneously by one process. The [getdtablesize(2)](/getdtablesize/2.md) system call returns the current system limit. ## RETURN VALUES If successful, **open**() and **openat**() return a non-negative integer, termed a file descriptor. They return -1 on failure, and set *errno* to indicate the error. ## ERRORS The named file is opened unless: - \[ENOTDIR\] A component of the path prefix is not a directory. - \[ENAMETOOLONG\] A component of a pathname exceeded 255 characters, or an entire path name exceeded 1023 characters. - \[ENOENT\] O\_CREAT is not set and the named file does not exist. - \[ENOENT\] A component of the path name that must exist does not exist. - \[EACCES\] Search permission is denied for a component of the path prefix. - \[EACCES\] The required permissions (for reading and/or writing) are denied for the given flags. - \[EACCES\] O\_TRUNC is specified and write permission is denied. - \[EACCES\] O\_CREAT is specified, the file does not exist, and the directory in which it is to be created does not permit writing. - \[EPERM\] O\_CREAT is specified, the file does not exist, and the directory in which it is to be created has its immutable flag set, see the [chflags(2)](/chflags/2.md) manual page for more information. - \[EPERM\] The named file has its immutable flag set and the file is to be modified. - \[EPERM\] The named file has its append-only flag set, the file is to be modified, and O\_TRUNC is specified or O\_APPEND is not specified. - \[ELOOP\] Too many symbolic links were encountered in translating the pathname. - \[EISDIR\] The named file is a directory, and the arguments specify it is to be modified. - \[EISDIR\] The named file is a directory, and the flags specified O\_CREAT without O\_DIRECTORY. - \[EROFS\] The named file resides on a read-only file system, and the file is to be modified. - \[EROFS\] O\_CREAT is specified and the named file would reside on a read-only file system. - \[EMFILE\] The process has already reached its limit for open file descriptors. - \[ENFILE\] The system file table is full. - \[EMLINK\] O\_NOFOLLOW was specified and the target is a symbolic link. POSIX specifies a different error for this case; see the note in STANDARDS below. - \[ENXIO\] The named file is a character special or block special file, and the device associated with this special file does not exist. - \[ENXIO\] O\_NONBLOCK is set, the named file is a fifo, O\_WRONLY is set, and no process has the file open for reading. - \[EINTR\] The **open**() operation was interrupted by a signal. - \[EOPNOTSUPP\] O\_SHLOCK or O\_EXLOCK is specified but the underlying file system does not support locking. - \[EOPNOTSUPP\] The named file is a special file mounted through a file system that does not support access to it (for example, NFS). - \[EWOULDBLOCK\] O\_NONBLOCK and one of O\_SHLOCK or O\_EXLOCK is specified and the file is locked. - \[ENOSPC\] O\_CREAT is specified, the file does not exist, and the directory in which the entry for the new file is being placed cannot be extended because there is no space left on the file system containing the directory. - \[ENOSPC\] O\_CREAT is specified, the file does not exist, and there are no free inodes on the file system on which the file is being created. - \[EDQUOT\] O\_CREAT is specified, the file does not exist, and the directory in which the entry for the new file is being placed cannot be extended because the user's quota of disk blocks on the file system containing the directory has been exhausted. - \[EDQUOT\] O\_CREAT is specified, the file does not exist, and the user's quota of inodes on the file system on which the file is being created has been exhausted. - \[EIO\] An I/O error occurred while making the directory entry or allocating the inode for O\_CREAT. - \[EINTEGRITY\] Corrupted data was detected while reading from the file system. - \[ETXTBSY\] The file is a pure procedure (shared text) file that is being executed and the **open**() system call requests write access. - \[EFAULT\] The *path* argument points outside the process's allocated address space. - \[EEXIST\] O\_CREAT and O\_EXCL were specified and the file exists. - \[EOPNOTSUPP\] An attempt was made to open a socket (not currently implemented). - \[EINVAL\] An attempt was made to open a descriptor with an illegal combination of O\_RDONLY, O\_WRONLY, or O\_RDWR, and O\_EXEC or O\_SEARCH. - \[EINVAL\] O\_CREAT is specified, and the last component of the *path* argument is invalid on the file system on which the file is being created. - \[EBADF\] The *path* argument does not specify an absolute path and the *fd* argument is neither AT\_FDCWD nor a valid file descriptor open for searching. - \[ENOTDIR\] The *path* argument is not an absolute path and *fd* is neither AT\_FDCWD nor a file descriptor associated with a directory. - \[ENOTDIR\] O\_DIRECTORY is specified and the file is not a directory. - \[ECAPMODE\] AT\_FDCWD is specified and the process is in capability mode. - \[ECAPMODE\] **open**() was called and the process is in capability mode. - \[ENOTCAPABLE\] *path* is an absolute path and the process is in capability mode. - \[ENOTCAPABLE\] *path* is an absolute path and O\_RESOLVE\_BENEATH is specified. - \[ENOTCAPABLE\] *path* contains a ".." component leading to a directory outside of the directory hierarchy specified by *fd* and the process is in capability mode. - \[ENOTCAPABLE\] *path* contains a ".." component leading to a directory outside of the directory hierarchy specified by *fd* and O\_RESOLVE\_BENEATH is specified. - \[ENOTCAPABLE\] *path* contains a ".." component, the vfs.lookup\_cap\_dotdot [sysctl(3)](/sysctl/3.md) is set, and the process is in capability mode. - \[ENOATTR\] O\_NAMEDATTR has been specified and the file object is not a named attribute directory or named attribute. ## SEE ALSO [chmod(2)](/chmod/2.md), [close(2)](/close/2.md), [dup(2)](/dup/2.md), [fexecve(2)](/fexecve/2.md), [fhopen(2)](/fhopen/2.md), [getdtablesize(2)](/getdtablesize/2.md), [getfh(2)](/getfh/2.md), [lgetfh(2)](/lgetfh/2.md), [lseek(2)](/lseek/2.md), [read(2)](/read/2.md), [umask(2)](/umask/2.md), [write(2)](/write/2.md), [fopen(3)](/fopen/3.md), [capsicum(4)](/capsicum/4.md), [unix(4)](/unix/4.md), [named\_attribute(7)](/named_attribute/7.md) ## STANDARDS These functions are specified by IEEE Std 1003.1-2008 (“POSIX.1”). FreeBSD sets *errno* to EMLINK instead of ELOOP as specified by POSIX when O\_NOFOLLOW is set in flags and the final component of pathname is a symbolic link to distinguish it from the case of too many symbolic link traversals in one of its non-final components. The Open Group Extended API Set 2 specification, that introduced the **\*at**() API, required that the test for whether *fd* is searchable is based on whether *fd* is open for searching, not whether the underlying directory currently permits searches. The present implementation of the *openat* system call is believed to be compatible with IEEE Std 1003.1-2008, 2017 Edition ("POSIX.1"), which specifies that behavior for O\_SEARCH, in the absence of the flag the implementation checks the current permissions of a directory. ## HISTORY The **open**() function appeared in Version 1 AT&T UNIX. The **openat**() function was introduced in FreeBSD 8.0. O\_DSYNC appeared in 13.0. O\_NAMEDATTR appeared in 15.0. O\_CLOFORK appeared in FreeBSD 15.0. ## BUGS The *mode* argument is variadic and may result in different calling conventions than might otherwise be expected.