xref: /openbsd-src/share/man/man9/fork1.9 (revision 43003dfe3ad45d1698bed8a37f2b0f5b14f20d4f)
1.\"	$OpenBSD: fork1.9,v 1.12 2009/07/28 12:27:03 jmc Exp $
2.\"	$NetBSD: fork1.9,v 1.3 1999/03/16 00:40:47 garbled Exp $
3.\"
4.\" Copyright (c) 1998 The NetBSD Foundation, Inc.
5.\" All rights reserved.
6.\"
7.\" This code is derived from software contributed to The NetBSD Foundation
8.\" by Jason R. Thorpe of the Numerical Aerospace Simulation Facility,
9.\" NASA Ames Research Center.
10.\"
11.\" Redistribution and use in source and binary forms, with or without
12.\" modification, are permitted provided that the following conditions
13.\" are met:
14.\" 1. Redistributions of source code must retain the above copyright
15.\"    notice, this list of conditions and the following disclaimer.
16.\" 2. Redistributions in binary form must reproduce the above copyright
17.\"    notice, this list of conditions and the following disclaimer in the
18.\"    documentation and/or other materials provided with the distribution.
19.\"
20.\" THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS
21.\" ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
22.\" TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
23.\" PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS
24.\" BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
25.\" CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
26.\" SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
27.\" INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
28.\" CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
29.\" ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
30.\" POSSIBILITY OF SUCH DAMAGE.
31.\"
32.Dd $Mdocdate: July 28 2009 $
33.Dt FORK1 9
34.Os
35.Sh NAME
36.Nm fork1
37.Nd create a new process
38.Sh SYNOPSIS
39.Fd #include <sys/types.h>
40.Fd #include <sys/proc.h>
41.Ft int
42.Fo "fork1"
43.Fa "struct proc *p1"
44.Fa "int exitsig"
45.Fa "int flags"
46.Fa "void *stack"
47.Fa "size_t stacksize"
48.Fa "void (*func)(void *)"
49.Fa "void *arg"
50.Fa "register_t *retval"
51.Fa "struct proc **rnewprocp"
52.Fc
53.Sh DESCRIPTION
54.Fn fork1
55creates a new process out of
56.Ar p1 ,
57which should be the current process.
58This function is used primarily to implement the
59.Xr fork 2 ,
60.Xr rfork 2 ,
61.Xr vfork 2
62system calls, as well as the
63.Xr kthread_create 9
64function.
65.Pp
66The
67.Ar flags
68argument is used to control the behavior of the fork and is created by
69a bitwise-OR of the following values:
70.Bl -tag -width FORK_SHAREFILES
71.It Dv FORK_FORK
72The call is done by the
73.Xr fork 2
74system call.
75Used only for statistics.
76.It Dv FORK_VFORK
77The call is done by the
78.Xr vfork 2
79system call.
80Used only for statistics.
81.It Dv FORK_RFORK
82The call is done by the
83.Xr rfork 2
84system call.
85Used only for statistics.
86.It Dv FORK_PPWAIT
87Suspend the parent process until the child is terminated (by calling
88.Xr _exit 2
89or abnormally), or makes a call to
90.Xr execve 2 .
91.It Dv FORK_SHAREFILES
92Let the child share the file descriptor table with the parent through
93fdshare().
94The default behavior is to copy the table through
95fdcopy().
96.It Dv FORK_CLEANFILES
97The child starts with a clean file descriptor table created by
98fdinit().
99.It Dv FORK_NOZOMBIE
100The child will be dissociated from the parent and will not leave a status
101for the parent to collect.
102See
103.Xr wait 2 .
104.It Dv FORK_SHAREVM
105The child will share the parent's address space.
106The default behavior is
107that the child gets a copy-on-write copy of the address space.
108.El
109.Pp
110If
111.Fa stack
112is not
113.Dv NULL ,
114the area starting at
115.Fa stack
116of extent
117.Fa stacksize
118will be used for the child's stack, instead of cloning the parent's
119stack.
120.Pp
121If
122.Fa retval
123is not
124.Dv NULL ,
125it will hold the following values after successful completion
126of the fork operation:
127.Bl -tag -width retval[0]
128.It Fa retval[0]
129This will contain the pid of the child process.
130.It Fa retval[1]
131In the parent process, this will contain the value 0.
132In the child process, this will contain 1.
133.El
134.Pp
135The signal
136.Fa exitsig
137is sent to the parent
138.Fa p1
139on exit of the new process.
140.Pp
141If
142.Fa func
143is not
144.Dv NULL ,
145the new process will begin execution by calling this function.
146It defaults to child_return, which returns to userland.
147.Pp
148If
149.Fa arg
150is not
151.Dv NULL ,
152it is the argument to the previous function.
153It defaults to a pointer to the new process.
154.Pp
155The newly created process is returned through
156.Fa *rnewprocp .
157.Sh RETURN VALUES
158Upon successful completion of the fork operation,
159.Fn fork1
160returns 0.
161Otherwise, the following error values are returned:
162.Bl -tag -width [EAGAIN]
163.It Bq Er EAGAIN
164The limit on the total number of system processes would be exceeded.
165.It Bq Er EAGAIN
166The limit
167.Dv RLIMIT_NPROC
168on the total number of processes under execution by this
169user id would be exceeded.
170.El
171.Sh SEE ALSO
172.Xr execve 2 ,
173.Xr fork 2 ,
174.Xr rfork 2 ,
175.Xr vfork 2 ,
176.Xr kthread_create 9 ,
177.Xr pfind 9 ,
178.Xr psignal 9 ,
179.Xr uvm_fork 9
180.Sh CAVEATS
181The
182.Nm
183function semantics are specific to
184.Ox .
185Other BSD systems have different semantics.
186.Pp
187The system never uses
188.Fn fork1
189with a non-null
190.Fa stack ,
191so that feature is not even tested.
192