1:mod:`fcntl` --- The ``fcntl`` and ``ioctl`` system calls
2=========================================================
3
4.. module:: fcntl
5   :platform: Unix
6   :synopsis: The fcntl() and ioctl() system calls.
7
8.. sectionauthor:: Jaap Vermeulen
9
10.. index::
11   pair: UNIX; file control
12   pair: UNIX; I/O control
13
14----------------
15
16This module performs file control and I/O control on file descriptors. It is an
17interface to the :c:func:`fcntl` and :c:func:`ioctl` Unix routines.  For a
18complete description of these calls, see :manpage:`fcntl(2)` and
19:manpage:`ioctl(2)` Unix manual pages.
20
21.. include:: ../includes/wasm-notavail.rst
22
23All functions in this module take a file descriptor *fd* as their first
24argument.  This can be an integer file descriptor, such as returned by
25``sys.stdin.fileno()``, or an :class:`io.IOBase` object, such as ``sys.stdin``
26itself, which provides a :meth:`~io.IOBase.fileno` that returns a genuine file
27descriptor.
28
29.. versionchanged:: 3.3
30   Operations in this module used to raise an :exc:`IOError` where they now
31   raise an :exc:`OSError`.
32
33.. versionchanged:: 3.8
34   The fcntl module now contains ``F_ADD_SEALS``, ``F_GET_SEALS``, and
35   ``F_SEAL_*`` constants for sealing of :func:`os.memfd_create` file
36   descriptors.
37
38.. versionchanged:: 3.9
39   On macOS, the fcntl module exposes the ``F_GETPATH`` constant, which obtains
40   the path of a file from a file descriptor.
41   On Linux(>=3.15), the fcntl module exposes the ``F_OFD_GETLK``, ``F_OFD_SETLK``
42   and ``F_OFD_SETLKW`` constants, which are used when working with open file
43   description locks.
44
45.. versionchanged:: 3.10
46   On Linux >= 2.6.11, the fcntl module exposes the ``F_GETPIPE_SZ`` and
47   ``F_SETPIPE_SZ`` constants, which allow to check and modify a pipe's size
48   respectively.
49
50.. versionchanged:: 3.11
51   On FreeBSD, the fcntl module exposes the ``F_DUP2FD`` and ``F_DUP2FD_CLOEXEC``
52   constants, which allow to duplicate a file descriptor, the latter setting
53   ``FD_CLOEXEC`` flag in addition.
54
55The module defines the following functions:
56
57
58.. function:: fcntl(fd, cmd, arg=0)
59
60   Perform the operation *cmd* on file descriptor *fd* (file objects providing
61   a :meth:`~io.IOBase.fileno` method are accepted as well).  The values used
62   for *cmd* are operating system dependent, and are available as constants
63   in the :mod:`fcntl` module, using the same names as used in the relevant C
64   header files. The argument *arg* can either be an integer value, or a
65   :class:`bytes` object. With an integer value, the return value of this
66   function is the integer return value of the C :c:func:`fcntl` call.  When
67   the argument is bytes it represents a binary structure, e.g. created by
68   :func:`struct.pack`. The binary data is copied to a buffer whose address is
69   passed to the C :c:func:`fcntl` call.  The return value after a successful
70   call is the contents of the buffer, converted to a :class:`bytes` object.
71   The length of the returned object will be the same as the length of the
72   *arg* argument. This is limited to 1024 bytes. If the information returned
73   in the buffer by the operating system is larger than 1024 bytes, this is
74   most likely to result in a segmentation violation or a more subtle data
75   corruption.
76
77   If the :c:func:`fcntl` fails, an :exc:`OSError` is raised.
78
79   .. audit-event:: fcntl.fcntl fd,cmd,arg fcntl.fcntl
80
81
82.. function:: ioctl(fd, request, arg=0, mutate_flag=True)
83
84   This function is identical to the :func:`~fcntl.fcntl` function, except
85   that the argument handling is even more complicated.
86
87   The *request* parameter is limited to values that can fit in 32-bits.
88   Additional constants of interest for use as the *request* argument can be
89   found in the :mod:`termios` module, under the same names as used in
90   the relevant C header files.
91
92   The parameter *arg* can be one of an integer, an object supporting the
93   read-only buffer interface (like :class:`bytes`) or an object supporting
94   the read-write buffer interface (like :class:`bytearray`).
95
96   In all but the last case, behaviour is as for the :func:`~fcntl.fcntl`
97   function.
98
99   If a mutable buffer is passed, then the behaviour is determined by the value of
100   the *mutate_flag* parameter.
101
102   If it is false, the buffer's mutability is ignored and behaviour is as for a
103   read-only buffer, except that the 1024 byte limit mentioned above is avoided --
104   so long as the buffer you pass is at least as long as what the operating system
105   wants to put there, things should work.
106
107   If *mutate_flag* is true (the default), then the buffer is (in effect) passed
108   to the underlying :func:`ioctl` system call, the latter's return code is
109   passed back to the calling Python, and the buffer's new contents reflect the
110   action of the :func:`ioctl`.  This is a slight simplification, because if the
111   supplied buffer is less than 1024 bytes long it is first copied into a static
112   buffer 1024 bytes long which is then passed to :func:`ioctl` and copied back
113   into the supplied buffer.
114
115   If the :c:func:`ioctl` fails, an :exc:`OSError` exception is raised.
116
117   An example::
118
119      >>> import array, fcntl, struct, termios, os
120      >>> os.getpgrp()
121      13341
122      >>> struct.unpack('h', fcntl.ioctl(0, termios.TIOCGPGRP, "  "))[0]
123      13341
124      >>> buf = array.array('h', [0])
125      >>> fcntl.ioctl(0, termios.TIOCGPGRP, buf, 1)
126      0
127      >>> buf
128      array('h', [13341])
129
130   .. audit-event:: fcntl.ioctl fd,request,arg fcntl.ioctl
131
132
133.. function:: flock(fd, operation)
134
135   Perform the lock operation *operation* on file descriptor *fd* (file objects providing
136   a :meth:`~io.IOBase.fileno` method are accepted as well). See the Unix manual
137   :manpage:`flock(2)` for details.  (On some systems, this function is emulated
138   using :c:func:`fcntl`.)
139
140   If the :c:func:`flock` fails, an :exc:`OSError` exception is raised.
141
142   .. audit-event:: fcntl.flock fd,operation fcntl.flock
143
144
145.. function:: lockf(fd, cmd, len=0, start=0, whence=0)
146
147   This is essentially a wrapper around the :func:`~fcntl.fcntl` locking calls.
148   *fd* is the file descriptor (file objects providing a :meth:`~io.IOBase.fileno`
149   method are accepted as well) of the file to lock or unlock, and *cmd*
150   is one of the following values:
151
152   * :const:`LOCK_UN` -- unlock
153   * :const:`LOCK_SH` -- acquire a shared lock
154   * :const:`LOCK_EX` -- acquire an exclusive lock
155
156   When *cmd* is :const:`LOCK_SH` or :const:`LOCK_EX`, it can also be
157   bitwise ORed with :const:`LOCK_NB` to avoid blocking on lock acquisition.
158   If :const:`LOCK_NB` is used and the lock cannot be acquired, an
159   :exc:`OSError` will be raised and the exception will have an *errno*
160   attribute set to :const:`EACCES` or :const:`EAGAIN` (depending on the
161   operating system; for portability, check for both values).  On at least some
162   systems, :const:`LOCK_EX` can only be used if the file descriptor refers to a
163   file opened for writing.
164
165   *len* is the number of bytes to lock, *start* is the byte offset at
166   which the lock starts, relative to *whence*, and *whence* is as with
167   :func:`io.IOBase.seek`, specifically:
168
169   * :const:`0` -- relative to the start of the file (:data:`os.SEEK_SET`)
170   * :const:`1` -- relative to the current buffer position (:data:`os.SEEK_CUR`)
171   * :const:`2` -- relative to the end of the file (:data:`os.SEEK_END`)
172
173   The default for *start* is 0, which means to start at the beginning of the file.
174   The default for *len* is 0 which means to lock to the end of the file.  The
175   default for *whence* is also 0.
176
177   .. audit-event:: fcntl.lockf fd,cmd,len,start,whence fcntl.lockf
178
179Examples (all on a SVR4 compliant system)::
180
181   import struct, fcntl, os
182
183   f = open(...)
184   rv = fcntl.fcntl(f, fcntl.F_SETFL, os.O_NDELAY)
185
186   lockdata = struct.pack('hhllhh', fcntl.F_WRLCK, 0, 0, 0, 0, 0)
187   rv = fcntl.fcntl(f, fcntl.F_SETLKW, lockdata)
188
189Note that in the first example the return value variable *rv* will hold an
190integer value; in the second example it will hold a :class:`bytes` object.  The
191structure lay-out for the *lockdata* variable is system dependent --- therefore
192using the :func:`flock` call may be better.
193
194
195.. seealso::
196
197   Module :mod:`os`
198      If the locking flags :data:`~os.O_SHLOCK` and :data:`~os.O_EXLOCK` are
199      present in the :mod:`os` module (on BSD only), the :func:`os.open`
200      function provides an alternative to the :func:`lockf` and :func:`flock`
201      functions.
202