.\"	$NetBSD: fcntl.2,v 1.50 2023/07/10 02:31:54 christos Exp $
.\"
.\" Copyright (c) 1983, 1993
.\"	The Regents of the University of California.  All rights reserved.
.\"
.\" Redistribution and use in source and binary forms, with or without
.\" modification, are permitted provided that the following conditions
.\" are met:
.\" 1. Redistributions of source code must retain the above copyright
.\"    notice, this list of conditions and the following disclaimer.
.\" 2. Redistributions in binary form must reproduce the above copyright
.\"    notice, this list of conditions and the following disclaimer in the
.\"    documentation and/or other materials provided with the distribution.
.\" 3. Neither the name of the University nor the names of its contributors
.\"    may be used to endorse or promote products derived from this software
.\"    without specific prior written permission.
.\"
.\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
.\" ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
.\" SUCH DAMAGE.
.\"
.\"     @(#)fcntl.2	8.2 (Berkeley) 1/12/94
.\"
.Dd July 5, 2023
.Dt FCNTL 2
.Os
.Sh NAME
.Nm fcntl
.Nd file descriptor control
.Sh LIBRARY
.Lb libc
.Sh SYNOPSIS
.In fcntl.h
.Ft int
.Fn fcntl "int fd" "int cmd" "..."
.Sh DESCRIPTION
.Fn fcntl
provides for control over descriptors.
The argument
.Fa fd
is a descriptor to be operated on by
.Fa cmd
as described below.
The third parameter is called
.Fa arg
and is technically a pointer to void, but it is
interpreted as an int by some commands and ignored by others.
.Pp
Commands are:
.Bl -tag -width F_DUPFD_CLOEXEC
.It Dv F_DUPFD
Return a new descriptor as follows:
.Pp
.Bl -bullet -compact -offset 4n
.It
Lowest numbered available descriptor greater than or equal to
.Fa arg ,
which is interpreted as an int.
.It
Same object references as the original descriptor.
.It
New descriptor shares the same file offset if the object
was a file.
.It
Same access mode (read, write or read/write).
.It
Same file status flags (i.e., both file descriptors
share the same file status flags).
.It
The close-on-exec flag associated with the new file descriptor
is cleared to remain open across
.Xr execve 2
system calls.
.El
.It Dv F_DUPFD_CLOEXEC
Same as
.Dv F_DUPFD ,
but sets the close-on-exec property on the file descriptor created.
.It Dv F_GETFD
Get the close-on-exec flag associated with the file descriptor
.Fa fd
as
.Dv FD_CLOEXEC .
If the returned value ANDed with
.Dv FD_CLOEXEC
is 0,
the file will remain open across
.Fn exec ,
otherwise the file will be closed upon execution of
.Fn exec
.Fa ( arg
is ignored).
.It Dv F_SETFD
Set the close-on-exec flag associated with
.Fa fd
to
.Fa arg ,
where
.Fa arg
is either 0 or
.Dv FD_CLOEXEC ,
as described above.
.It Dv F_GETFL
Get descriptor status flags, as described below
.Fa ( arg
is ignored).
.It Dv F_SETFL
Set descriptor status flags to
.Fa arg ,
which is interpreted as an int.
.It Dv F_GETOWN
Get the process ID or process group
currently receiving
.Dv SIGIO
and
.Dv SIGURG
signals; process groups are returned
as negative values
.Fa ( arg
is ignored).
.It Dv F_SETOWN
Set the process or process group
to receive
.Dv SIGIO
and
.Dv SIGURG
signals;
process groups are specified by supplying
.Fa arg
as negative, otherwise
.Fa arg
is interpreted as a process ID.
The argument
.Fa arg
is interpreted as an int.
.It Dv F_CLOSEM
Close all file descriptors greater than or equal to
.Fa fd .
.It Dv F_MAXFD
Return the maximum file descriptor number currently open by the process.
.It Dv F_GETNOSIGPIPE
Return if the
.Dv O_NOSIGPIPE
flag is set in the file descriptor.
.It Dv F_SETNOSIGPIPE
Set or clear the
.Dv O_NOSIGPIPE
in the file descriptor.
.It Dv F_GETPATH
Place a pathname corresponding to
.Fa fd
in the buffer pointed to by
.Fa arg .
.Fa arg
should be pointing to a buffer of at least
.Dv MAXPATHLEN .
.It Dv F_ADD_SEALS
Add seals specified in
.Fa arg
to
.Fa fd
to restrict possible operations on
.Fa fd
as described below.
Like flags, multiple seals can be specified at once.
Additionally, specifying seals that are already associated with
.Fa fd
is a no-op.
.It Dv F_GET_SEALS
Get the seals currently associated with
.Fa fd
as described below
.Fa ( arg
is ignored).
.El
.Pp
The set of valid flags for the
.Dv F_GETFL
and
.Dv F_SETFL
flags are as follows:
.Dv O_APPEND ,
.Dv O_ASYNC ,
.Dv O_SYNC ,
.Dv O_NONBLOCK ,
.Dv O_DSYNC ,
.Dv O_RSYNC ,
.Dv O_ALT_IO ,
.Dv O_DIRECT ,
.Dv O_NOSIGPIPE .
These flags are described in
.Xr open 2 .
.Pp
Several commands are available for doing advisory file locking;
they all operate on the following structure:
.Bd -literal
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 */
};
.Ed
.Pp
The commands available for advisory record locking are as follows:
.Bl -tag -width F_SETLKWX
.It Dv F_GETLK
Get the first lock that blocks the lock description pointed to by the
third argument,
.Fa arg ,
taken as a pointer to a
.Fa "struct flock"
(see above).
The information retrieved overwrites the information passed to
.Nm
in the
.Fa 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
.Fa l_type ,
which is set to
.Dv F_UNLCK .
.It Dv F_SETLK
Set or clear a file segment lock according to the lock description
pointed to by the third argument,
.Fa arg ,
taken as a pointer to a
.Fa "struct flock"
(see above).
As specified by the value of
.Fa l_type ,
.Dv F_SETLK
is used to establish shared (or read) locks
.Pq Dv F_RDLCK
or exclusive (or write) locks,
.Pq Dv F_WRLCK ,
as well as remove either type of lock
.Pq Dv F_UNLCK .
If a shared or exclusive lock cannot be set,
.Nm
returns immediately with
.Er EAGAIN .
.It Dv F_SETLKW
This command is the same as
.Dv 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
.Nm
is waiting for a region, the
.Nm
will be interrupted if the signal handler has not specified the
.Dv SA_RESTART
(see
.Xr sigaction 2 ) .
.El
.Pp
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.
.Pp
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.
.Pp
The value of
.Fa l_whence
is
.Dv SEEK_SET ,
.Dv SEEK_CUR ,
or
.Dv SEEK_END
to indicate that the relative offset,
.Fa l_start
bytes, will be measured from the start of the file,
current position, or end of the file, respectively.
The value of
.Fa l_len
is the number of consecutive bytes to be locked.
If
.Fa l_len
is negative, the result is undefined.
The
.Fa l_pid
field is only used with
.Dv F_GETLK
to return the process ID of the process holding a blocking lock.
After a successful
.Dv F_GETLK
request, the value of
.Fa l_whence
is
.Dv SEEK_SET .
.Pp
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
.Fa l_len
is set to zero.
If
.Fa l_whence
and
.Fa l_start
point to the beginning of the file, and
.Fa l_len
is zero, the entire file is locked.
If an application wishes only to do entire file locking, the
.Xr flock 2
system call is much more efficient.
.Pp
There is at most one type of lock set for each byte in the file.
Before a successful return from an
.Dv F_SETLK
or an
.Dv 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
.Dv F_SETLK
or an
.Dv 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.
.Pp
Possible seals are:
.Bl -tag -width F_SEAL_FUTURE_WRITE
.It Dv F_SEAL_SEAL
Prevent any further seals from being added to
.Fa fd .
.It Dv F_SEAL_SHRINK
Prevent the size of
.Fa fd
from decreasing.
.It Dv F_SEAL_GROW
Prevent the size of
.Fa fd
from increasing.
.It Dv F_SEAL_WRITE
Prevent any write operations to
.Fa fd .
.Dv F_SEAL_WRITE
cannot be applied if
.Fa fd
has any memory mappings.
.It Dv F_SEAL_FUTURE_WRITE
Like
.Dv F_SEAL_WRITE
but allow any current memory mappings of
.Fa fd
to remain open, including those with
.Dv PROT_WRITE .
.El
.Sh NOTES
For
.Dv F_GETPATH :
.Bl -bullet -compact
.It
For vnodes, functionality is implemented using the reverse
.Xr namei 9
cache.
The implications of this are
.Bl -bullet -compact
.It
For hard links where the file descriptor can resolve to multiple pathnames,
the first entry found in the cache is returned.
.It
.Dv F_GETPATH
may fail if the corresponding entry has been evicted from the LRU
.Xr namei 9
cache and return
.Er ENOENT .
.El
.It
For a file descriptor created by
.Xr memfd_create 2 ,
the name provided at
.Fa fd
creation, with the prefix
.Dq memfd:
is used.
.It
Other types of file descriptors are not handled, as well as symbolic
links since there is currently no way to obtain a file descriptor
pointing to a symbolic link.
.El
.Sh RETURN VALUES
Upon successful completion, the value returned depends on
.Fa cmd
as follows:
.Bl -tag -width F_GET_SEALS -offset indent
.It Dv F_DUPFD
A new file descriptor.
.It Dv F_GETFD
Value of flag (only the low-order bit is defined).
.It Dv F_GETFL
Value of flags.
.It Dv F_GETOWN
Value of file descriptor owner.
.It Dv F_MAXFD
Value of the highest file descriptor open by the process.
.It Dv F_GET_SEALS
Value of the seals currently associated with
.Fa fd .
.It other
Value other than \-1.
.El
.Pp
Otherwise, a value of \-1 is returned and
.Va errno
is set to indicate the error.
.Sh COMPATIBILITY
This interface follows the completely stupid semantics of
.At V
and
.St -p1003.1-88
that require that all locks associated with a file for a given process are
removed when \fIany\fP 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
.Xr getpwnam 3
to retrieve a record,
the lock will be lost because
.Xr getpwnam 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.
.Pp
Another minor semantic problem with this interface is that
locks are not inherited by a child process created using the
.Xr fork 2
function.
The
.Xr flock 2
interface has much more rational last close semantics and
allows locks to be inherited by child processes.
Calling
.Xr 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
.Xr flock 2
and
.Nm
locks may be safely used concurrently.
.Pp
All locks associated with a file for a given process are
removed when the process terminates.
.Pp
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
.Er EDEADLK
error.
.Sh ERRORS
.Fn fcntl
will fail if:
.Bl -tag -width Er
.It Bq Er EACCES
The argument
.Fa cmd
is
.Dv F_GETPATH
and read or search permission was denied for a component of the pathname.
.It Bq Er EAGAIN
The argument
.Fa arg
is
.Dv F_SETLK ,
the type of lock
.Pq Fa l_type
is a shared lock
.Pq Dv F_RDLCK
or exclusive lock
.Pq Dv 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.
.It Bq Er EBADF
.Fa fildes
is not a valid open file descriptor.
.Pp
The argument
.Fa cmd
is
.Dv F_SETLK
or
.Dv F_SETLKW ,
the type of lock
.Pq Fa l_type
is a shared lock
.Pq Dv F_RDLCK ,
and
.Fa fildes
is not a valid file descriptor open for reading.
.Pp
The argument
.Fa cmd
is
.Dv F_SETLK
or
.Dv F_SETLKW ,
the type of lock
.Pq Fa l_type
is an exclusive lock
.Pq Dv F_WRLCK ,
and
.Fa fildes
is not a valid file descriptor open for writing.
.It Bq Er EBUSY
The argument
.Fa cmd
is
.Dv F_ADD_SEALS ,
.Fa arg
contains
.Dv F_SEAL_WRITE
and
.Fa fd
is currently mapped by
.Xr mmap 2 .
.It Bq Er EDEADLK
The argument
.Fa cmd
is
.Dv F_SETLKW ,
and a deadlock condition was detected.
.It Bq Er EINTR
The argument
.Fa cmd
is
.Dv F_SETLKW ,
and the function was interrupted by a signal.
.It Bq Er EINVAL
The argument
.Fa cmd
is invalid.
.Pp
The argument
.Fa cmd
is
.Dv F_DUPFD
and
.Fa arg
is negative or greater than the maximum allowable number
(see
.Xr getdtablesize 3 ) .
.Pp
The argument
.Fa cmd
is
.Dv F_GETLK ,
.Dv F_SETLK ,
or
.Dv F_SETLKW
and the data to which
.Fa arg
points is not valid, or
.Fa fildes
refers to a file that does not support locking.
.Pp
The argument
.Fa cmd
is
.Dv F_ADD_SEALS
or
.Dv F_GET_SEALS
and
.Fa fd
does not support seals.
.Pp
The argument
.Fa cmd
is
.Dv F_ADD_SEALS
and
.Fa arg
contains set bits for unsupported seals.
.It Bq Er EMFILE
The argument
.Fa cmd
is
.Dv 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
.Fa arg
are available.
.It Bq Er ENFILE
.Fa cmd
is
.Dv F_DUPFD
and system-wide the maximum allowed number of file descriptors are
currently open.
.It Bq Er ENOENT
The argument
.Fa cmd
is
.Dv F_GETPATH
and a component of the pathname no longer exists.
.It Bq Er ENOLCK
The argument
.Fa cmd
is
.Dv F_SETLK
or
.Dv 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.
.It Bq Er ENOMEM
The argument
.Fa cmd
is
.Dv F_GETPATH
and insufficient memory is available.
.Pp
The argument
.Fa cmd
is
.Dv F_GETLK ,
.Dv F_SETLK ,
or
.Dv F_SETLKW ,
and the file lock limit for the current unprivileged user
has been reached.
It can be modified using the
.Li kern.maxfiles
.Xr sysctl 7 .
.It Bq Er EPERM
The argument
.Fa cmd
is
.Dv F_ADD_SEALS
and
.Fa fd
already has
.Dv F_SEAL_SEAL .
.It Bq Er ERANGE
The argument
.Fa cmd
is
.Dv F_GETPATH
and the resulting path would be greater than
.Dv MAXPATHLEN .
.It Bq Er ESRCH
.Fa cmd
is
.Dv F_SETOWN
and
the process ID given as argument is not in use.
.El
.Sh SEE ALSO
.Xr close 2 ,
.Xr execve 2 ,
.Xr flock 2 ,
.Xr open 2 ,
.Xr sigaction 2 ,
.Xr getdtablesize 3
.Sh STANDARDS
The
.Fn fcntl
function conforms to
.St -p1003.1-90 .
.Sh HISTORY
The
.Fn fcntl
function call appeared in
.Bx 4.2 .