1.\" $NetBSD: vfork.2,v 1.27 2014/07/18 16:02:50 dholland Exp $ 2.\" 3.\" Copyright (c) 1980, 1991, 1993 4.\" The Regents of the University of California. All rights reserved. 5.\" 6.\" Redistribution and use in source and binary forms, with or without 7.\" modification, are permitted provided that the following conditions 8.\" are met: 9.\" 1. Redistributions of source code must retain the above copyright 10.\" notice, this list of conditions and the following disclaimer. 11.\" 2. Redistributions in binary form must reproduce the above copyright 12.\" notice, this list of conditions and the following disclaimer in the 13.\" documentation and/or other materials provided with the distribution. 14.\" 3. Neither the name of the University nor the names of its contributors 15.\" may be used to endorse or promote products derived from this software 16.\" without specific prior written permission. 17.\" 18.\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 19.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 20.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 21.\" ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 22.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 23.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 24.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 25.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 26.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 27.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 28.\" SUCH DAMAGE. 29.\" 30.\" @(#)vfork.2 8.1 (Berkeley) 6/4/93 31.\" 32.Dd July 18, 2014 33.Dt VFORK 2 34.Os 35.Sh NAME 36.Nm vfork 37.Nd spawn new process in a virtual memory efficient way 38.Sh LIBRARY 39.Lb libc 40.Sh SYNOPSIS 41.In unistd.h 42.Ft pid_t 43.Fn vfork void 44.Sh DESCRIPTION 45The 46.Nm 47system call creates a new process that does not have a new 48virtual address space, but rather shares address space with the 49parent, thus avoiding potentially expensive copy-on-write operations 50normally associated with creating a new process. 51It is useful when the purpose of 52.Xr fork 2 53would have been to create a new system context for an 54.Xr execve 2 . 55The 56.Nm 57system call differs from 58.Xr fork 2 59in that the child borrows the parent's memory and thread of 60control until a call to 61.Xr execve 2 62or an exit (either by a call to 63.Xr _exit 2 64or abnormally). 65The parent process is suspended while the child is using its resources. 66.Pp 67The 68.Nm 69system call returns 0 in the child's context and (later) the pid 70of the child in the parent's context. 71.Pp 72The 73.Nm 74system call can normally be used just like 75.Xr fork 2 . 76It does not work, however, to return while running in the child's context 77from the procedure that called 78.Fn vfork 79since the eventual return from 80.Fn vfork 81would then return to a no longer existing stack frame. 82Be careful, also, to call 83.Xr _exit 2 84rather than 85.Xr exit 3 86if you can't 87.Xr execve 2 , 88since 89.Xr exit 3 90will flush and close standard I/O channels, and thereby mess up the 91standard I/O data structures 92in the parent process. 93(Even with 94.Xr fork 2 95it is wrong to call 96.Xr exit 3 97since buffered data would then be flushed twice.) 98.Sh RETURN VALUES 99Same as for 100.Xr fork 2 . 101.Sh ERRORS 102Same as for 103.Xr fork 2 . 104.Sh SEE ALSO 105.Xr execve 2 , 106.Xr fork 2 , 107.Xr sigaction 2 , 108.Xr wait 2 109.Sh HISTORY 110The 111.Fn vfork 112function call appeared in 113.Bx 3.0 . 114In 115.Bx 4.4 , 116the semantics were changed to only suspend the parent and not share 117the address space. 118The original semantics were reintroduced in 119.Nx 1.4 . 120.Sh BUGS 121Portable applications should not depend on the memory sharing 122semantics of 123.Fn vfork 124as implementations exist that implement 125.Fn vfork 126as plain 127.Xr fork 2 . 128.Pp 129To avoid a possible deadlock situation, processes that are children 130in the middle of a 131.Fn vfork 132are never sent 133.Dv SIGTTOU 134or 135.Dv SIGTTIN 136signals; rather, output or 137.Xr ioctl 2 138calls are allowed and input attempts result in an end-of-file indication. 139