xref: /openbsd/lib/libc/sys/execve.2 (revision 3cab2bb3)
1.\"	$OpenBSD: execve.2,v 1.54 2020/04/10 07:31:45 jca Exp $
2.\"	$NetBSD: execve.2,v 1.9 1995/02/27 12:32:25 cgd Exp $
3.\"
4.\" Copyright (c) 1980, 1991, 1993
5.\"	The Regents of the University of California.  All rights reserved.
6.\"
7.\" Redistribution and use in source and binary forms, with or without
8.\" modification, are permitted provided that the following conditions
9.\" are met:
10.\" 1. Redistributions of source code must retain the above copyright
11.\"    notice, this list of conditions and the following disclaimer.
12.\" 2. Redistributions in binary form must reproduce the above copyright
13.\"    notice, this list of conditions and the following disclaimer in the
14.\"    documentation and/or other materials provided with the distribution.
15.\" 3. Neither the name of the University nor the names of its contributors
16.\"    may be used to endorse or promote products derived from this software
17.\"    without specific prior written permission.
18.\"
19.\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
20.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
21.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
22.\" ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
23.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
24.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
25.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
26.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
27.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
28.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
29.\" SUCH DAMAGE.
30.\"
31.\"     @(#)execve.2	8.3 (Berkeley) 1/24/94
32.\"
33.Dd $Mdocdate: April 10 2020 $
34.Dt EXECVE 2
35.Os
36.Sh NAME
37.Nm execve
38.Nd execute a file
39.Sh SYNOPSIS
40.In unistd.h
41.Ft int
42.Fn execve "const char *path" "char *const argv[]" "char *const envp[]"
43.Sh DESCRIPTION
44.Fn execve
45transforms the calling process into a new process.
46The new process is constructed from an ordinary file,
47whose name is pointed to by
48.Fa path ,
49called the
50.Em new process file .
51This file is either an executable object file,
52or a file of data for an interpreter.
53An executable object file consists of an identifying header,
54followed by pages of data representing the initial program (text)
55and initialized data pages.
56Additional pages may be specified by the header to be initialized
57with zero data; see
58.Xr elf 5 .
59.Pp
60An interpreter file begins with a line of the form:
61.Bd -filled -offset indent
62.Sy #!\&
63.Em interpreter
64.Bq Em arg
65.Ed
66.Pp
67When an interpreter file is passed to
68.Fn execve
69the system instead calls
70.Fn execve
71with 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 original
78.Ar path
79becomes the second argument;
80otherwise,
81.Ar path
82becomes the first argument.
83The original arguments are shifted over to become the subsequent arguments.
84The zeroth argument, normally the name of the file being executed, is left
85unchanged.
86.Pp
87The argument
88.Fa argv
89is a pointer to a null-terminated array of
90character pointers to NUL-terminated character strings.
91These strings construct the argument list to be made available to the new
92process.
93At least one non-null argument must be present in the array;
94by custom, the first element should be
95the name of the executed program (for example, the last component of
96.Fa path ) .
97.Pp
98The argument
99.Fa envp
100is also a pointer to a null-terminated array of
101character pointers to NUL-terminated strings.
102A pointer to this array is normally stored in the global variable
103.Va environ .
104These strings pass information to the
105new process that is not directly an argument to the command (see
106.Xr environ 7 ) .
107.Pp
108File descriptors open in the calling process image remain open in
109the new process image, except for those for which the close-on-exec
110flag is set (see
111.Xr close 2
112and
113.Xr fcntl 2 ) .
114Descriptors that remain open are unaffected by
115.Fn execve .
116In the case of a new setuid or setgid executable being executed, if
117file descriptors 0, 1, or 2 (representing stdin, stdout, and stderr)
118are currently unallocated, these descriptors will be opened to point to
119some system file like
120.Pa /dev/null .
121The intent is to ensure these descriptors are not unallocated, since
122many libraries make assumptions about the use of these 3 file descriptors.
123.Pp
124Signals set to be ignored in the calling process,
125with the exception of
126.Dv SIGCHLD ,
127are set to be ignored in
128the
129new process.
130Other signals
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 ) .
154The set-user-ID and set-group-ID bits have no effect if the
155new process image file is located on a file system mounted with
156the nosuid flag.
157The process will be started without the new permissions.
158.Pp
159The new process also inherits the following attributes from
160the calling process:
161.Pp
162.Bl -tag -width parent_process_ID -offset indent -compact
163.It process ID
164see
165.Xr getpid 2
166.It parent process ID
167see
168.Xr getppid 2
169.It process group ID
170see
171.Xr getpgrp 2
172.It session ID
173see
174.Xr getsid 2
175.It access groups
176see
177.Xr getgroups 2
178.It working directory
179see
180.Xr chdir 2
181.It root directory
182see
183.Xr chroot 2
184.It control terminal
185see
186.Xr termios 4
187.It resource usages
188see
189.Xr getrusage 2
190.It interval timers
191see
192.Xr getitimer 2
193(unless process image file is setuid or setgid,
194in which case all timers are disabled)
195.It resource limits
196see
197.Xr getrlimit 2
198.It file mode mask
199see
200.Xr umask 2
201.It signal mask
202see
203.Xr sigaction 2 ,
204.Xr sigprocmask 2
205.El
206.Pp
207When a program is executed as a result of an
208.Fn execve
209call, it is entered as follows:
210.Bd -literal -offset indent
211main(int argc, char **argv, char **envp)
212.Ed
213.Pp
214where
215.Fa argc
216is the number of elements in
217.Fa argv
218(the
219.Dq arg count )
220and
221.Fa argv
222points to the array of character pointers
223to the arguments themselves.
224.Sh RETURN VALUES
225As the
226.Fn execve
227function overlays the current process image
228with a new process image the successful call
229has no process to return to.
230If
231.Fn execve
232does return to the calling process an error has occurred; the
233return value will be \-1 and the global variable
234.Va errno
235is set to indicate the error.
236.Sh ERRORS
237.Fn execve
238will fail and return to the calling process if:
239.Bl -tag -width Er
240.It Bq Er ENOTDIR
241A component of the path prefix is not a directory.
242.It Bq Er ENAMETOOLONG
243A component of a pathname exceeded
244.Dv NAME_MAX
245characters, or an entire pathname (including the terminating NUL)
246exceeded
247.Dv PATH_MAX
248bytes.
249.It Bq Er ENOENT
250The new process file does not exist.
251.It Bq Er ELOOP
252Too many symbolic links were encountered in translating the pathname.
253.It Bq Er EACCES
254Search permission is denied for a component of the path prefix.
255.It Bq Er EACCES
256The new process file is not an ordinary file.
257.It Bq Er EACCES
258The new process file mode denies execute permission.
259.It Bq Er EACCES
260The new process file is on a filesystem mounted with execution
261disabled
262.Pf ( Dv MNT_NOEXEC
263in
264.In sys/mount.h ) .
265.It Bq Er EACCES
266The new process file is marked with
267.Xr ld 1
268.Fl z Cm wxneeded
269to perform W^X violating operations, but it is located on a file
270system not allowing such operations, being mounted without the
271.Xr mount 8
272.Fl o Cm wxallowed
273flag.
274.It Bq Er EACCES
275The parent used
276.Xr pledge 2
277to declare an
278.Va execpromise ,
279and that is not permitted for setuid or setgid images.
280.It Bq Er ENOEXEC
281The new process file has the appropriate access
282permission, but has an invalid magic number in its header.
283.It Bq Er ETXTBSY
284The new process file is a pure procedure (shared text)
285file that is currently open for writing by some process.
286.It Bq Er ENOMEM
287The new process requires more virtual memory than
288is allowed by the imposed maximum
289.Pq Xr getrlimit 2 .
290.It Bq Er E2BIG
291The number of bytes in the new process's argument list
292is larger than the system-imposed limit.
293The limit in the system as released is 524288 bytes
294.Pf ( Dv ARG_MAX ) .
295.It Bq Er EFAULT
296The new process file is not as long as indicated by
297the size values in its header.
298.It Bq Er EFAULT
299.Fa path ,
300.Fa argv ,
301or
302.Fa envp
303point
304to an illegal address.
305.It Bq Er EINVAL
306.Fa argv
307did not contain at least one element.
308.It Bq Er EIO
309An I/O error occurred while reading from the file system.
310.It Bq Er ENFILE
311During startup of an
312.Em interpreter ,
313the system file table was found to be full.
314.El
315.Sh SEE ALSO
316.Xr _exit 2 ,
317.Xr fork 2 ,
318.Xr execl 3 ,
319.Xr exit 3 ,
320.Xr elf 5 ,
321.Xr environ 7
322.Sh STANDARDS
323The
324.Fn execve
325function is expected to conform to
326.St -p1003.1-2008 .
327.Sh HISTORY
328The predecessor of these functions, the former
329.Fn exec
330system call, first appeared in
331.At v1 .
332The
333.Fn execve
334function first appeared in
335.At v7 .
336.Sh CAVEATS
337If a program is
338.Em setuid
339to a non-superuser, but is executed when the real
340.Em uid
341is
342.Dq root ,
343then the process has some of the powers of a superuser as well.
344.Pp
345.St -p1003.1-2008
346permits
347.Nm
348to leave
349.Dv SIGCHLD
350as ignored in the new process; portable programs cannot rely on
351.Nm
352resetting it to the default disposition.
353