EXECVE(2) System Calls Manual EXECVE(2)

execve, fexecve — execute a file

Standard C Library (libc, -lc)

#include <unistd.h>

int
execve(const char *path, char *const argv[], char *const envp[]);

int
fexecve(int fd, char *const argv[], char *const envp[]);

The () system call transforms the calling process into a new process. The new process is constructed from an ordinary file, whose name is pointed to by path, called the . The fexecve() system call is equivalent to execve() except that the file to be executed is determined by the file descriptor fd instead of a path. This file is either an executable object file, or a file of data for an interpreter. An executable object file consists of an identifying header, followed by pages of data representing the initial program (text) and initialized data pages. Additional pages may be specified by the header to be initialized with zero data; see elf(5) and a.out(5).

An interpreter file begins with a line of the form:

#!interpreter [arg]

When an interpreter file is execve'd, the system actually execve's the specified interpreter. If the optional arg is specified, it becomes the first argument to the interpreter, and the name of the originally execve'd file becomes the second argument; otherwise, the name of the originally execve'd file becomes the first argument. The original arguments are shifted over to become the subsequent arguments. The zeroth argument, normally the name of the execve'd file, is left unchanged. The interpreter named by interpreter must not itself be an interpreter file. (See script(7) for a detailed discussion of interpreter file execution.)

The argument argv is a pointer to a null-terminated array of character pointers to null-terminated character strings. These strings construct the argument list to be made available to the new process. At least one argument must be present in the array; by custom, the first element should be the name of the executed program (for example, the last component of path).

The argument envp is also a pointer to a null-terminated array of character pointers to null-terminated strings. A pointer to this array is normally stored in the global variable environ. These strings pass information to the new process that is not directly an argument to the command (see environ(7)).

File descriptors open in the calling process image remain open in the new process image, except for those for which the close-on-exec flag is set (see close(2) and fcntl(2)) which are closed. Descriptors that remain open are unaffected by (), except that the close-on-fork flag is cleared from all descriptors remaining open.

In the case of a new setuid or setgid executable being executed, if file descriptors 0, 1, or 2 (representing , , and ) are currently unallocated, these descriptors will be opened to point to some system file like /dev/null. The intent is to ensure these descriptors are not unallocated, since many libraries make assumptions about the use of these three file descriptors.

Signals set to be ignored in the calling process are set to be ignored in the new process. Signals which are set to be caught in the calling process image are set to default action in the new process image. Blocked signals remain blocked regardless of changes to the signal action. The signal stack is reset to be undefined (see sigaction(2) for more information).

If the set-user-ID mode bit of the new process image file is set (see chmod(2)), the effective user ID of the new process image is set to the owner ID of the new process image file. If the set-group-ID mode bit of the new process image file is set, the effective group ID of the new process image is set to the group ID of the new process image file. (The effective group ID is the first element of the group list.) The real user ID, real group ID and other group IDs of the new process image remain the same as the calling process image. After any set-user-ID and set-group-ID processing, the effective user ID is recorded as the saved set-user-ID, and the effective group ID is recorded as the saved set-group-ID. These values may be used in changing the effective IDs later (see setuid(2)). The set-ID bits are not honored if the respective file system has the nosuid option enabled or if the new process file is an interpreter file. Syscall tracing is disabled if effective IDs are changed.

The new process also inherits the following attributes from the calling process:

process ID see getpid(2)
parent process ID see getppid(2)
process group ID see getpgrp(2)
access groups see getgroups(2)
working directory see chdir(2)
root directory see chroot(2)
control terminal see termios(4)
resource usages see getrusage(2)
interval timers see getitimer(2)
resource limits see getrlimit(2)
file mode mask see umask(2)
signal mask see sigaction(2), sigprocmask(2)

When a program is executed as a result of an () system call, it is entered as follows:

main(argc, argv, envp)
int argc;
char **argv, **envp;

where argc (the “arg count”) is the number of elements in argv, and argv points to the array of character pointers to the arguments themselves.

The () function ignores the file offset of fd. Since execute permission is checked by fexecve(), the file descriptor fd need not have been opened with the O_EXEC flag. However, if the file to be executed denies read permission for the process preparing to do the exec, the only way to provide the fd to fexecve() is to use the O_EXEC flag when opening fd. Note that the file to be executed can not be open for writing.

As the execve() system call overlays the current process image with a new process image the successful call has no process to return to. If execve() does return to the calling process an error has occurred; the return value will be -1 and the global variable errno is set to indicate the error.

The execve() system call will fail and return to the calling process if:

[]
The number of bytes in the new process' argument list is larger than the system-imposed limit. The default compile time limit is 262144 bytes and is specified in the variable NCARGS in <sys/param.h> and get be read from the sysctl(3) MIB variable KERN_ARGMAX.
[]
Search permission is denied for a component of the path prefix, the new process file is not an ordinary file, its file mode denies execute permission, or it is on a file system mounted with execution disabled (MNT_NOEXEC in <sys/mount.h>).
[]
A setuid(7) process has exceeded the current resource limit for the number of processes it is allowed to run concurrently.
[]
The new process file is not as long as indicated by the size values in its header; or the path, argv, or envp arguments point to an illegal address.
[]
An I/O error occurred while reading from the file system.
[]
Too many symbolic links were encountered in translating the pathname.
[]
A component of a pathname exceeded {NAME_MAX} characters, or an entire path name exceeded {PATH_MAX} characters.
[]
The new process file does not exist, or the new process file is a script starting with ‘#!’ and the script interpreter does not exist.
[]
The new process file has the appropriate access permission, but has an invalid magic number in its header.
[]
The new process requires more virtual memory than is allowed by the imposed maximum (getrlimit(2)).
[]
A component of the path prefix is not a directory.
[]
The new process file is a pure procedure (shared text) file that is currently open for writing or reading by some process.

In addition, the fexecve() will fail and return to the calling process if:

[]
The fd argument is not a valid file descriptor open for executing.

_exit(2), fcntl(2), fork(2), open(2), execl(3), exit(3), sysctl(3), a.out(5), elf(5), environ(7), script(7), mount(8)

The execve() system call conforms to IEEE Std 1003.1-2001 (“POSIX.1”). with the exception of reopening descriptors 0, 1, and/or 2 in certain circumstances. A future update of the Standard is expected to require this behavior, and it may become the default for non-privileged processes as well. The support for executing interpreted programs is an extension. The fexecve() system call conforms to The Open Group Extended API Set 2 specification.

The execve() function call first appeared in Version 7 AT&T UNIX. The fexecve() system call appeared in NetBSD 10.0.

If a program is to a non-super-user, but is executed when the real is “root”, then the program has some of the powers of a super-user as well.

July 8, 2025 NetBSD 11.0