xref: /netbsd-src/lib/libc/sys/execve.2 (revision 23c8222edbfb0f0932d88a8351d3a0cf817dfb9e)
1.\"	$NetBSD: execve.2,v 1.33 2004/05/13 10:20:58 wiz 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.\"     @(#)execve.2	8.5 (Berkeley) 6/1/94
31.\"
32.Dd August 11, 2002
33.Dt EXECVE 2
34.Os
35.Sh NAME
36.Nm execve
37.Nd execute a file
38.Sh LIBRARY
39.Lb libc
40.Sh SYNOPSIS
41.In unistd.h
42.Ft int
43.Fn execve "const char *path" "char *const argv[]" "char *const envp[]"
44.Sh DESCRIPTION
45.Fn execve
46transforms the calling process into a new process.
47The new process is constructed from an ordinary file,
48whose name is pointed to by
49.Fa path ,
50called the
51.Em new process file .
52This file is either an executable object file,
53or a file of data for an interpreter.
54An executable object file consists of an identifying header,
55followed by pages of data representing the initial program (text)
56and initialized data pages.
57Additional pages may be specified
58by the header to be initialized with zero data;  see
59.Xr a.out 5 .
60.Pp
61An interpreter file begins with a line of the form:
62.Pp
63.Bd -filled -offset indent -compact
64.Sy \&#!
65.Em interpreter
66.Bq Em arg
67.Ed
68.Pp
69When an interpreter file is
70.Fn execve Ns d
71the system runs the specified
72.Em interpreter .
73If the optional
74.Em arg
75is specified, it becomes the first argument to the
76.Em interpreter ,
77and the name of the originally
78.Fn execve Ns d
79file becomes the second argument;
80otherwise, the name of the originally
81.Fn execve Ns d
82file becomes the first argument.
83The original arguments are shifted over to become the subsequent arguments.
84The zeroth argument, normally the name of the
85.Fn execve Ns d
86file, is left unchanged.
87The interpreter named by
88.Em interpreter
89must not itself be an interpreter file.
90.Pp
91The argument
92.Fa argv
93is a pointer to a null-terminated array of
94character pointers to null-terminated character strings.
95These strings construct the argument list to be made available to the new
96process.
97By custom, the first element should be
98the name of the executed program (for example, the last component of
99.Fa path ) .
100.Pp
101The argument
102.Fa envp
103is also a pointer to a null-terminated array of
104character pointers to null-terminated strings.
105A pointer to this array is normally stored in the global variable
106.Va environ .
107These strings pass information to the
108new process that is not directly an argument to the command (see
109.Xr environ 7 ) .
110.Pp
111File descriptors open in the calling process image remain open in
112the new process image, except for those for which the close-on-exec
113flag is set (see
114.Xr close 2
115and
116.Xr fcntl 2 ) .
117Descriptors that remain open are unaffected by
118.Fn execve .
119.Pp
120In the case of a new setuid or setgid executable being executed, if
121file descriptors 0, 1, or 2 (representing stdin, stdout, and stderr)
122are currently unallocated, these descriptors will be opened to point to
123some system file like
124.Pa /dev/null .
125The intent is to ensure these descriptors are not unallocated, since
126many libraries make assumptions about the use of these 3 file descriptors.
127.Pp
128Signals set to be ignored in the calling process are set to be ignored in
129the new process.
130Signals which are set to be caught in the calling process image
131are set to default action in the new process image.
132Blocked signals remain blocked regardless of changes to the signal action.
133The signal stack is reset to be undefined (see
134.Xr sigaction 2
135for more information).
136.Pp
137If the set-user-ID mode bit of the new process image file is set
138(see
139.Xr chmod 2 ) ,
140the effective user ID of the new process image is set to the owner ID
141of the new process image file.
142If the set-group-ID mode bit of the new process image file is set,
143the effective group ID of the new process image is set to the group ID
144of the new process image file.
145(The effective group ID is the first element of the group list.)
146The real user ID, real group ID and
147other group IDs of the new process image remain the same as the calling
148process image.
149After any set-user-ID and set-group-ID processing,
150the effective user ID is recorded as the saved set-user-ID,
151and the effective group ID is recorded as the saved set-group-ID.
152These values may be used in changing the effective IDs later (see
153.Xr setuid 2 ) .
154.ne 1i
155.Pp
156The new process also inherits the following attributes from
157the calling process:
158.Pp
159.Bl -column parent_process_ID -offset indent -compact
160.It process ID Ta see Xr getpid 2
161.It parent process ID Ta see Xr getppid 2
162.It process group ID Ta see Xr getpgrp 2
163.It access groups Ta see Xr getgroups 2
164.It working directory Ta see Xr chdir 2
165.It root directory Ta see Xr chroot 2
166.It control terminal Ta see Xr termios 4
167.It resource usages Ta see Xr getrusage 2
168.It interval timers Ta see Xr getitimer 2
169.It resource limits Ta see Xr getrlimit 2
170.It file mode mask Ta see Xr umask 2
171.It signal mask Ta see Xr sigaction 2 ,
172.Xr sigprocmask 2
173.El
174.Pp
175When a program is executed as a result of an
176.Fn execve
177call, it is entered as follows:
178.Bd -literal -offset indent
179main(argc, argv, envp)
180int argc;
181char **argv, **envp;
182.Ed
183.Pp
184where
185.Fa argc
186is the number of elements in
187.Fa argv
188(the
189.Dq arg count )
190and
191.Fa argv
192points to the array of character pointers
193to the arguments themselves.
194.Sh RETURN VALUES
195As the
196.Fn execve
197function overlays the current process image
198with a new process image the successful call
199has no process to return to.
200If
201.Fn execve
202does return to the calling process an error has occurred; the
203return value will be \-1 and the global variable
204.Va errno
205is set to indicate the error.
206.Sh ERRORS
207.Fn execve
208will fail and return to the calling process if:
209.Bl -tag -width Er
210.It Bq Er ENOTDIR
211A component of the path prefix is not a directory.
212.It Bq Er ENAMETOOLONG
213A component of a pathname exceeded
214.Dv {NAME_MAX}
215characters, or an entire path name exceeded
216.Dv {PATH_MAX}
217characters.
218.It Bq Er ENOENT
219The new process file does not exist.
220.It Bq Er ELOOP
221Too many symbolic links were encountered in translating the pathname.
222.It Bq Er EACCES
223Search permission is denied for a component of the path prefix,
224the new process file is not an ordinary file,
225its file mode denies execute permission, or
226it is on a filesystem mounted with execution
227disabled
228.Pf ( Dv MNT_NOEXEC
229in
230.Ao Pa sys/mount.h Ac ) .
231.It Bq Er ENOEXEC
232The new process file has the appropriate access
233permission, but has an invalid magic number in its header.
234.It Bq Er ETXTBSY
235The new process file is a pure procedure (shared text)
236file that is currently open for writing or reading by some process.
237.ne 1i
238.It Bq Er ENOMEM
239The new process requires more virtual memory than
240is allowed by the imposed maximum
241.Pq Xr getrlimit 2 .
242.It Bq Er E2BIG
243The number of bytes in the new process's argument list
244is larger than the system-imposed limit.
245The limit in the system as released is 262144 bytes
246.Pf ( Dv NCARGS
247in
248.Ao Pa sys/param.h Ac ) .
249.It Bq Er EFAULT
250The new process file is not as long as indicated by
251the size values in its header.
252.It Bq Er EFAULT
253.Fa path ,
254.Fa argv ,
255or
256.Fa envp
257point
258to an illegal address.
259.It Bq Er EIO
260An I/O error occurred while reading from the file system.
261.El
262.Sh SEE ALSO
263.Xr _exit 2 ,
264.Xr fork 2 ,
265.Xr execl 3 ,
266.Xr environ 7
267.Sh STANDARDS
268The
269.Fn execve
270function conforms to
271.St -p1003.1-90 .
272.Sh HISTORY
273The
274.Fn execve
275function call appeared in
276.Bx 4.2 .
277.Sh BUGS
278If a program is
279.Em setuid
280to a non-super-user, but is executed when
281the real
282.Em uid
283is
284.Dq root ,
285then the program has some of the powers of a super-user as well.
286