Cameron Katri's Manual Page Server

Manual Page Search Parameters
FCNTL(2) System Calls Manual FCNTL(2)

fcntl — file control

#include <fcntl.h>

int
fcntl(int fildes, int cmd, ...);

() provides for control over descriptors. The argument fildes is a descriptor to be operated on by cmd as follows:

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 associated with the new file descriptor is cleared so that the descriptor remains open across an execv(2) system call.
Like F_DUPFD, except that the close-on-exec flag associated with the new file descriptor is set.
Get the flags associated with the file descriptor fildes, as described below (arg is ignored).
Set the file descriptor flags to arg.
Get descriptor status flags, as described below (arg is ignored).
Set descriptor status flags to arg.
Get the process ID or process group currently receiving SIGIO and SIGURG signals; process groups are returned as negative values (arg is ignored).
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.
Get the path of the file descriptor
Get the non firmlinked path of the file descriptor Fildes. The argument must be a buffer of size or greater.
Preallocate file storage space. Note: upon success, the space that is allocated can be the size requested, larger than the size requested, or (if the F_ALLOCATEALL flag is not provided) smaller than the space requested.
Deallocate a region and replace it with a hole. Subsequent reads of the affected region will return bytes of zeros that are usually not backed by physical blocks. This will not change the actual file size. Holes must be aligned to file system block boundaries. This will fail on file systems that do not support this interface.
Deprecated. In previous releases, this would allow a process with root privileges to truncate a file without zeroing space. For security reasons, this operation is no longer supported and will instead truncate the file in the same manner as truncate(2).
Issue an advisory read async with no copy to user.
Turn read ahead off/on. A zero value in arg disables read ahead. A non-zero value in arg turns read ahead on.
Turns data caching off/on. A non-zero value in arg turns data caching off. A value of zero in arg turns data caching on.
Get disk device information. Currently this only returns the disk device address that corresponds to the current file offset. Note that the system may return -1 as the disk device address if the file is not backed by physical blocks. This is subject to change.
Variant of F_LOG2PHYS that uses the passed in file offset and length.
Does the same thing as fsync(2) then issues a barrier command to the drive (arg is ignored). The barrier applies to I/O that have been flushed with fsync(2) on the same device before. These operations are guaranteed to be persisted before any other I/O that would follow the barrier, although no assumption should be made on what has been persisted or not when this call returns. After the barrier has been issued, operations on other FDs that have been fsync'd before can still be re-ordered by the device, but not after the barrier. This is typically useful to guarantee valid state on disk when ordering is a concern but durability is not. A barrier can be used to order two phases of operations on a set of file descriptors and ensure that no file can possibly get persisted with the effect of the second phase without the effect of the first one. To do so, execute operations of phase one, then fsync(2) each FD and issue a single barrier. Finally execute operations of phase two. This is currently implemented on HFS and APFS. It requires hardware support, which Apple SSDs are guaranteed to provide.
Does the same thing as fsync(2) then asks the drive to flush all buffered data to the permanent storage device (arg is ignored). As this drains the entire queue of the device and acts as a barrier, data that had been fsync'd on the same device before is guaranteed to be persisted when this call returns. This is currently implemented on HFS, MS-DOS (FAT), Universal Disk Format (UDF) and APFS file systems. The operation may take quite a while to complete. Certain FireWire drives have also been known to ignore the request to flush their buffered data.
Determines whether a SIGPIPE signal will be generated when a write fails on a pipe or socket for which there is no reader. If arg is non-zero, SIGPIPE generation is disabled for descriptor fildes, while an arg of zero enables it (the default).
Returns whether a SIGPIPE signal will be generated when a write fails on a pipe or socket for which there is no reader. The semantics of the return value match those of the arg of F_SETNOSIGPIPE.
Transfer any extra space in the file past the logical EOF (as previously allocated via F_PREALLOCATE) to another file. The other file is specified via a file descriptor as the lone extra argument. Both descriptors must reference regular files in the same volume.

The flags for the F_GETFD and F_SETFD commands are as follows:

Close-on-exec; the given file descriptor will be automatically closed in the successor process image when one of the execv(2) or posix_spawn(2) family of system calls is invoked.

The flags for the F_GETFL and F_SETFL commands are as follows:

Non-blocking I/O; if no data is available to a read(2) call, or if a write(2) operation would block, the read or write call returns -1 with the error EAGAIN.
Force each write to append at the end of file; corresponds to the O_APPEND flag of open(2).
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.

Several commands 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 */
        };

The commands available for advisory record locking are as follows:

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 function call except for the lock type which is set to F_UNLCK. If a lock that does not support the discovery of lock ownership by process (one created by the flock(2) system call or the open(2) system call with the O_SHLOCK or O_EXLOCK flag) is found, l_pid is set to -1.
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.
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)).

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, the result is undefined. The l_pid field is only used with F_GETLK to return the process ID of the process holding a blocking lock. 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) 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.

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 getpwname(3) to retrieve a record, the lock will be lost because getpwname(3) 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) function. The flock(2) interface has much more rational last close semantics and allows locks to be inherited by child processes. Flock(2) 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. Note that flock(2) and fcntl locks may be safely used concurrently.

All locks associated with a file for a given process are removed when the process terminates.

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.

The F_PREALLOCATE command operates on the following structure:

        typedef struct fstore {
	    u_int32_t fst_flags;      /* IN: flags word */
	    int       fst_posmode;    /* IN: indicates offset field */
	    off_t     fst_offset;     /* IN: start of the region */
	    off_t     fst_length;     /* IN: size of the region */
	    off_t     fst_bytesalloc; /* OUT: number of bytes allocated */
        } fstore_t;

The flags (fst_flags) for the F_PREALLOCATE command are as follows:

Allocate contiguous space. (Note that the file system may ignore this request if fst_length is very large.)
Allocate all requested space or no space at all.
Allocate space that is not freed when close(2) is called. (Note that the file system may ignore this request.)

The position modes (fst_posmode) for the F_PREALLOCATE command indicate how to use the offset field. The modes are as follows:

Allocate from the physical end of file. In this case, fst_length indicates the number of newly allocated bytes desired.
Allocate from the volume offset.

The F_PUNCHHOLE command operates on the following structure:

        typedef struct fpunchhole {
            u_int32_t fp_flags;     /* unused */
            u_int32_t reserved;     /* (to maintain 8-byte alignment) */
            off_t     fp_offset;    /* IN: start of the region */
            off_t     fp_length;    /* IN: size of the region */
        } fpunchhole_t;

The F_RDADVISE command operates on the following structure which holds information passed from the user to the system:

        struct radvisory {
           off_t   ra_offset;  /* offset into the file */
           int     ra_count;   /* size of the read     */
        };

The F_LOG2PHYS command operates on the following structure:

        struct log2phys {
            u_int32_t l2p_flags;        /* unused so far */
            off_t     l2p_contigbytes;  /* unused so far */
            off_t     l2p_devoffset;    /* bytes into device */
        };

The F_LOG2PHYS_EXT command operates on the same structure as F_LOG2PHYS but treats it as an in/out:

        struct log2phys {
            u_int32_t l2p_flags;        /* unused so far */
            off_t     l2p_contigbytes;  /* IN: number of bytes to be queried;
                                           OUT: number of contiguous bytes allocated at this position */
            off_t     l2p_devoffset;    /* IN: bytes into file;
                                           OUT: bytes into device */
        };

If fildes is a socket, then the F_SETNOSIGPIPE and F_GETNOSIGPIPE commands are directly analogous, and fully interoperate with the SO_NOSIGPIPE option of setsockopt(2) and getsockopt(2) respectively.

Upon successful completion, the value returned depends on cmd as follows:

A new file descriptor.
Value of flag (only the low-order bit is defined).
Value of flags.
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.

The fcntl() system call will fail if:

[]
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.
[]
The argument cmd is F_SETSIZE and the calling process does not have root privileges.
[]
Fildes is not a valid open file descriptor.

The argument cmd is F_SETLK or F_SETLKW, the type of lock (l_type) is a shared lock (F_RDLCK), and fildes 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 fildes is not a valid file descriptor open for writing.

The argument cmd is F_PREALLOCATE and the calling process does not have file write permission.

The argument cmd is F_LOG2PHYS or F_LOG2PHYS_EXT and fildes is not a valid file descriptor open for reading.

The argument cmd is F_TRANSFEREXTENTS and either file descriptor does not correspond to a valid regular file, or either file is not open for writing.

[]
The argument cmd is F_SETLKW, and a deadlock condition was detected.
[]
The argument cmd is F_PREALLOCATE, F_PEOFPOSMODE is set and preallocating fst_length bytes on fildes would exceed the maximum file size.
[]
The argument cmd is F_SETLKW, and the function was interrupted by a signal.
[]
Cmd is F_DUPFD and arg is negative or greater than the maximum allowable number (see getdtablesize(2)).

The argument cmd is F_GETLK, F_SETLK, or F_SETLKW and the data to which arg points is not valid, or fildes refers to a file that does not support locking.

The argument cmd is F_PREALLOCATE and the fst_posmode is not a valid mode, or when F_PEOFPOSMODE is set and fst_offset is a non-zero value, or when F_VOLPOSMODE is set and fst_offset is a negative or zero value.

The argument cmd is F_PUNCHHOLE and either fp_offset or fp_length are negative, or when both fp_offset and fp_length are not multiples of the file system block size, or when either fp_flags or reserved is non-zero value.

The argument cmd is F_TRANSFEREXTENTS and the additional file descriptor is negative or both file descriptors reference the same file.

[]
Cmd is F_DUPFD and the maximum allowed number of file descriptors are currently open.
[]
The argument cmd is F_DUPED 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.
[]
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.
[]
The argument cmd is F_PREALLOCATE and either there is no space available on the volume containing fildes or fst_flags contains F_ALLOCATEALL and there is not enough space available on the volume containing fildes to satisfy the entire request.

The argument cmd is F_PUNCHHOLE and there is not enough space available on the volume containing fildes to satisfy the request. As an example, a filesystem that supports cloned files may return this error if punching a hole requires the creation of a clone and there is not enough space available to do so.

[]
A return value would overflow its representation. For example, cmd is F_GETLK, F_SETLK, or F_SETLKW and the smallest (or, if l_len is non-zero, the largest) offset of a byte in the requested segment will not fit in an object of type off_t.
[]
The argument cmd is F_PUNCHHOLE and the calling process does not have file write permission.
[]
Cmd is F_SETOWN and the process ID given as argument is not in use.
[]
Cmd is F_TRANSFEREXTENTS and the given files aren't on an APFS volume.
[]
Cmd is F_TRANSFEREXTENTS and the referenced files are not in the same volume.

close(2), execve(2), flock(2), getdtablesize(2), open(2), pipe(2), setsockopt(2), socket(2), sigaction(3)

The fcntl() function call appeared in 4.2BSD.

August 12, 2021 BSD 4.2