--- name: "fcntl" 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/fcntl" --- # fcntl(2) file control ## NAME **fcntl** – file control ## LIBRARY Standard C Library (libc, -lc) ## SYNOPSIS **#include \** *int*\ **fcntl**(*int fd*, *int cmd*, *...*); ## DESCRIPTION The **fcntl**() system call provides for control over descriptors. The argument *fd* is a descriptor to be operated on by *cmd* as described below. Depending on the value of *cmd*, **fcntl**() can take an additional third argument *arg*. Unless otherwise noted below for a specific operation, *arg* has type *int*. - F\_DUPFD Return a new descriptor as follows: - Lowest numbered available descriptor greater than or equal to *arg*. - Same object references as the original descriptor. - New descriptor shares the same file offset if the object was a file. - Same access mode (read, write or read/write). - Same file status flags (i.e., both file descriptors share the same file status flags). - The close-on-exec flag FD\_CLOEXEC associated with the new file descriptor is cleared, so the file descriptor is to remain open across [execve(2)](/execve/2.md) system calls. - The fork-on-exec flag FD\_CLOFORK associated with the new file descriptor is cleared, so the file descriptor is to remain open across [fork(2)](/fork/2.md) system calls. - The FD\_RESOLVE\_BENEATH flag, described below, will be set if it was set on the original descriptor. - F\_DUPFD\_CLOEXEC Like F\_DUPFD, but the FD\_CLOEXEC flag associated with the new file descriptor is set, so the file descriptor is closed when [execve(2)](/execve/2.md) system call executes. - F\_DUPFD\_CLOFORK Like F\_DUPFD, but the FD\_CLOFORK flag associated with the new file descriptor is set, so the file descriptor is closed when [fork(2)](/fork/2.md) system call executes. - F\_DUP2FD It is functionally equivalent to ``` dup2(fd, arg) ``` - F\_DUP2FD\_CLOEXEC Like F\_DUP2FD, but the FD\_CLOEXEC flag associated with the new file descriptor is set. The F\_DUP2FD and F\_DUP2FD\_CLOEXEC constants are not portable, so they should not be used if portability is needed. Use **dup2**() instead of F\_DUP2FD. - F\_DUP3FD Used to implement the **dup3**() call. Do not use it. - F\_GETFD Get the flags associated with the file descriptor *fd*. The following flags are defined: - FD\_CLOEXEC The file will be closed upon execution of **exec**() *(*, *arg* is ignored). Otherwise, the file descriptor will remain open. - FD\_CLOFORK The file will be closed upon execution of the **fork**() family of system calls. - FD\_RESOLVE\_BENEATH All path name lookups relative to that file descriptor will behave as if the lookup had O\_RESOLVE\_BENEATH or AT\_RESOLVE\_BENEATH semantics. It is not permitted to call [fchdir(2)](/fchdir/2.md) or [fchroot(2)](/fchroot/2.md) on such a file descriptor. The FD\_RESOLVE\_BENEATH flag is sticky, meaning that it is preserved by [dup(2)](/dup/2.md) and similar operations, and opening a directory with [openat(2)](/openat/2.md) where the directory descriptor has the flag set causes the new directory descriptor to also have the flag set. A file descriptor with the FD\_RESOLVE\_BENEATH set cannot be used as either the source or target descriptor in [renameat(2)](/renameat/2.md) or [renameat2(2)](/renameat2/2.md) system calls. - F\_SETFD Set flags associated with *fd* to *arg*. The available flags are FD\_CLOEXEC, FD\_CLOFORK and FD\_RESOLVE\_BENEATH. The FD\_RESOLVE\_BENEATH flag cannot be cleared once set. - F\_GETFL Get descriptor status flags, as described below *(*, *arg* is ignored). - F\_SETFL Set descriptor status flags to *arg*. - F\_GETOWN Get the process ID or process group currently receiving SIGIO and SIGURG signals; process groups are returned as negative values *(*, *arg* is ignored). - F\_SETOWN Set the process or process group to receive SIGIO and SIGURG signals; process groups are specified by supplying *arg* as negative, otherwise *arg* is interpreted as a process ID. - F\_READAHEAD Set or clear the read ahead amount for sequential access to the third argument, *arg*, which is rounded up to the nearest block size. A zero value in *arg* turns off read ahead, a negative value restores the system default. - F\_RDAHEAD Equivalent to Darwin counterpart which sets read ahead amount of 128KB when the third argument, *arg* is non-zero. A zero value in *arg* turns off read ahead. - F\_ADD\_SEALS Add seals to the file as described below, if the underlying filesystem supports seals. - F\_GET\_SEALS Get seals associated with the file, if the underlying filesystem supports seals. - F\_ISUNIONSTACK Check if the vnode is part of a union stack (either the "union" flag from [mount(2)](/mount/2.md) or unionfs). This is a hack not intended to be used outside of libc. - F\_KINFO Fills a *struct* *kinfo\_file* for the file referenced by the specified file descriptor. The *arg* argument should point to the storage for *struct* *kinfo\_file*. The *kf\_structsize* member of the passed structure must be initialized with the sizeof of *struct* *kinfo\_file*, to allow for the interface versioning and evolution. The flags for the F\_GETFL and F\_SETFL commands are as follows: - O\_NONBLOCK Non-blocking I/O; if no data is available to a [read(2)](/read/2.md) system call, or if a [write(2)](/write/2.md) operation would block, the read or write call returns -1 with the error EAGAIN. - O\_APPEND Force each write to append at the end of file; corresponds to the O\_APPEND flag of [open(2)](/open/2.md). - O\_DIRECT 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. - O\_ASYNC Enable the SIGIO signal to be sent to the process group when I/O is possible, e.g., upon availability of data to be read. - O\_SYNC Enable synchronous writes. Corresponds to the O\_SYNC flag of [open(2)](/open/2.md). O\_FSYNC is an historical synonym for O\_SYNC. - O\_DSYNC Enable synchronous data writes. Corresponds to the O\_DSYNC flag of [open(2)](/open/2.md). The seals that may be applied with F\_ADD\_SEALS are as follows: - F\_SEAL\_SEAL Prevent any further seals from being applied to the file. - F\_SEAL\_SHRINK Prevent the file from being shrunk with [ftruncate(2)](/ftruncate/2.md). - F\_SEAL\_GROW Prevent the file from being enlarged with [ftruncate(2)](/ftruncate/2.md). - F\_SEAL\_WRITE Prevent any further [write(2)](/write/2.md) calls to the file. Any writes in progress will finish before **fcntl**() returns. If any writeable mappings exist, F\_ADD\_SEALS will fail and return EBUSY. Seals are on a per-inode basis and require support by the underlying filesystem. If the underlying filesystem does not support seals, F\_ADD\_SEALS and F\_GET\_SEALS will fail and return EINVAL. Several operations are available for doing advisory file locking; they all operate on the following structure: ``` struct flock { off_t l_start; /* starting offset */ off_t l_len; /* len = 0 means until end of file */ pid_t l_pid; /* lock owner */ short l_type; /* lock type: read/write, etc. */ short l_whence; /* type of l_start */ int l_sysid; /* remote system id or zero for local */ }; ``` These advisory file locking operations take a pointer to *struct* *flock* as the third argument *arg*. The commands available for advisory record locking are as follows: - F\_GETLK Get the first lock that blocks the lock description pointed to by the third argument, *arg*, taken as a pointer to a *struct flock* (see above). The information retrieved overwrites the information passed to **fcntl**() in the *flock* structure. If no lock is found that would prevent this lock from being created, the structure is left unchanged by this system call except for the lock type which is set to F\_UNLCK. - F\_SETLK Set or clear a file segment lock according to the lock description pointed to by the third argument, *arg*, taken as a pointer to a *struct flock* (see above). F\_SETLK is used to establish shared (or read) locks (F\_RDLCK) or exclusive (or write) locks, (F\_WRLCK), as well as remove either type of lock (F\_UNLCK). If a shared or exclusive lock cannot be set, **fcntl**() returns immediately with EAGAIN. - F\_SETLKW This command is the same as F\_SETLK except that if a shared or exclusive lock is blocked by other locks, the process waits until the request can be satisfied. If a signal that is to be caught is received while **fcntl**() is waiting for a region, the **fcntl**() will be interrupted if the signal handler has not specified the SA\_RESTART (see [sigaction(2)](/sigaction/2.md)). When a shared lock has been set on a segment of a file, other processes can set shared locks on that segment or a portion of it. A shared lock prevents any other process from setting an exclusive lock on any portion of the protected area. A request for a shared lock fails if the file descriptor was not opened with read access. An exclusive lock prevents any other process from setting a shared lock or an exclusive lock on any portion of the protected area. A request for an exclusive lock fails if the file was not opened with write access. The value of *l\_whence* is SEEK\_SET, SEEK\_CUR, or SEEK\_END to indicate that the relative offset, *l\_start* bytes, will be measured from the start of the file, current position, or end of the file, respectively. The value of *l\_len* is the number of consecutive bytes to be locked. If *l\_len* is negative, *l\_start* means end edge of the region. The *l\_pid* and *l\_sysid* fields are only used with F\_GETLK to return the process ID of the process holding a blocking lock and the system ID of the system that owns that process. Locks created by the local system will have a system ID of zero. After a successful F\_GETLK request, the value of *l\_whence* is SEEK\_SET. Locks may start and extend beyond the current end of a file, but may not start or extend before the beginning of the file. A lock is set to extend to the largest possible value of the file offset for that file if *l\_len* is set to zero. If *l\_whence* and *l\_start* point to the beginning of the file, and *l\_len* is zero, the entire file is locked. If an application wishes only to do entire file locking, the [flock(2)](/flock/2.md) system call is much more efficient. There is at most one type of lock set for each byte in the file. Before a successful return from an F\_SETLK or an F\_SETLKW request when the calling process has previously existing locks on bytes in the region specified by the request, the previous lock type for each byte in the specified region is replaced by the new lock type. As specified above under the descriptions of shared locks and exclusive locks, an F\_SETLK or an F\_SETLKW request fails or blocks respectively when another process has existing locks on bytes in the specified region and the type of any of those locks conflicts with the type specified in the request. The queuing for F\_SETLKW requests on local files is fair; that is, while the thread is blocked, subsequent requests conflicting with its requests will not be granted, even if these requests do not conflict with existing locks. This interface follows the completely stupid semantics of System V and IEEE Std 1003.1-1988 (“POSIX.1”) that require that all locks associated with a file for a given process are removed when *any* file descriptor for that file is closed by that process. This semantic means that applications must be aware of any files that a subroutine library may access. For example if an application for updating the password file locks the password file database while making the update, and then calls [getpwnam(3)](/getpwnam/3.md) to retrieve a record, the lock will be lost because [getpwnam(3)](/getpwnam/3.md) opens, reads, and closes the password database. The database close will release all locks that the process has associated with the database, even if the library routine never requested a lock on the database. Another minor semantic problem with this interface is that locks are not inherited by a child process created using the [fork(2)](/fork/2.md) system call. The [flock(2)](/flock/2.md) interface has much more rational last close semantics and allows locks to be inherited by child processes. The [flock(2)](/flock/2.md) system call is recommended for applications that want to ensure the integrity of their locks when using library routines or wish to pass locks to their children. The **fcntl**(), [flock(2)](/flock/2.md), and [lockf(3)](/lockf/3.md) locks are compatible. Processes using different locking interfaces can cooperate over the same file safely. However, only one of such interfaces should be used within the same process. If a file is locked by a process through [flock(2)](/flock/2.md), any record within the file will be seen as locked from the viewpoint of another process using **fcntl**() or [lockf(3)](/lockf/3.md), and vice versa. Note that **fcntl**(*F\_GETLK*) returns -1 in *l\_pid* if the process holding a blocking lock previously locked the file descriptor by [flock(2)](/flock/2.md). All locks associated with a file for a given process are removed when the process terminates. All locks obtained before a call to [execve(2)](/execve/2.md) remain in effect until the new program releases them. If the new program does not know about the locks, they will not be released until the program exits. A potential for deadlock occurs if a process controlling a locked region is put to sleep by attempting to lock the locked region of another process. This implementation detects that sleeping until a locked region is unlocked would cause a deadlock and fails with an EDEADLK error. ## RETURN VALUES Upon successful completion, the value returned depends on *cmd* as follows: - F\_DUPFD A new file descriptor. - F\_DUP2FD A file descriptor equal to *arg*. - F\_GETFD Value of flags. - F\_GETFL Value of flags. - F\_GETOWN Value of file descriptor owner. - other Value other than -1. Otherwise, a value of -1 is returned and *errno* is set to indicate the error. ## ERRORS The **fcntl**() system call will fail if: - \[EAGAIN\] The argument *cmd* is F\_SETLK, the type of lock (*l\_type*) is a shared lock (F\_RDLCK) or exclusive lock (F\_WRLCK), and the segment of a file to be locked is already exclusive-locked by another process; or the type is an exclusive lock and some portion of the segment of a file to be locked is already shared-locked or exclusive-locked by another process. - \[EBADF\] The *fd* argument is not a valid open file descriptor. The argument *cmd* is F\_DUP2FD, and *arg* is not a valid file descriptor. The argument *cmd* is F\_SETLK or F\_SETLKW, the type of lock (*l\_type*) is a shared lock (F\_RDLCK), and *fd* is not a valid file descriptor open for reading. The argument *cmd* is F\_SETLK or F\_SETLKW, the type of lock (*l\_type*) is an exclusive lock (F\_WRLCK), and *fd* is not a valid file descriptor open for writing. - \[EBUSY\] The argument *cmd* is F\_ADD\_SEALS, attempting to set F\_SEAL\_WRITE, and writeable mappings of the file exist. - \[EDEADLK\] The argument *cmd* is F\_SETLKW, and a deadlock condition was detected. - \[EINTR\] The argument *cmd* is F\_SETLKW, and the system call was interrupted by a signal. - \[EINVAL\] The *cmd* argument is F\_DUPFD and *arg* is negative or greater than the maximum allowable number (see [getdtablesize(2)](/getdtablesize/2.md)). The argument *cmd* is F\_GETLK, F\_SETLK or F\_SETLKW and the data to which *arg* points is not valid. The argument *cmd* is F\_ADD\_SEALS or F\_GET\_SEALS, and the underlying filesystem does not support sealing. The argument *cmd* is invalid. - \[EMFILE\] The argument *cmd* is F\_DUPFD and the maximum number of file descriptors permitted for the process are already in use, or no file descriptors greater than or equal to *arg* are available. - \[ENOTTY\] The *fd* argument is not a valid file descriptor for the requested operation. This may be the case if *fd* is a device node, or a descriptor returned by [kqueue(2)](/kqueue/2.md). - \[ENOLCK\] The argument *cmd* is F\_SETLK or F\_SETLKW, and satisfying the lock or unlock request would result in the number of locked regions in the system exceeding a system-imposed limit. - \[EOPNOTSUPP\] The argument *cmd* is F\_GETLK, F\_SETLK or F\_SETLKW and *fd* refers to a file for which locking is not supported. - \[EOVERFLOW\] The argument *cmd* is F\_GETLK, F\_SETLK or F\_SETLKW and an *off\_t* calculation overflowed. - \[EPERM\] The *cmd* argument is F\_SETOWN and the process ID or process group given as an argument is in a different session than the caller. The *cmd* argument is F\_ADD\_SEALS and the F\_SEAL\_SEAL seal has already been set. - \[ESRCH\] The *cmd* argument is F\_SETOWN and the process ID given as argument is not in use. In addition, if *fd* refers to a descriptor open on a terminal device (as opposed to a descriptor open on a socket), a *cmd* of F\_SETOWN can fail for the same reasons as in [tcsetpgrp(3)](/tcsetpgrp/3.md), and a *cmd* of F\_GETOWN for the reasons as stated in [tcgetpgrp(3)](/tcgetpgrp/3.md). ## SEE ALSO [close(2)](/close/2.md), [dup2(2)](/dup2/2.md), [execve(2)](/execve/2.md), [flock(2)](/flock/2.md), [getdtablesize(2)](/getdtablesize/2.md), [open(2)](/open/2.md), [sigaction(2)](/sigaction/2.md), [lockf(3)](/lockf/3.md), [tcgetpgrp(3)](/tcgetpgrp/3.md), [tcsetpgrp(3)](/tcsetpgrp/3.md) ## STANDARDS The F\_DUP2FD and F\_DUP3FD constants are not portable. They are provided for compatibility with AIX and Solaris. Per Version 4 of the Single UNIX Specification (“SUSv4”), a call with F\_SETLKW should fail with \[EINTR\] after any caught signal and should continue waiting during thread suspension such as a stop signal. However, in this implementation a call with F\_SETLKW is restarted after catching a signal with a SA\_RESTART handler or a thread suspension such as a stop signal. ## HISTORY The **fcntl**() system call appeared in 4.2BSD. The F\_DUP2FD constant first appeared in FreeBSD 7.1. The F\_DUPFD\_CLOFORK and F\_DUP3FD flags appeared in FreeBSD 15.0.