1.\" 2.\" Copyright (c) 2000-2010 Apple Inc. All rights reserved. 3.\" 4.\" @APPLE_OSREFERENCE_LICENSE_HEADER_START@ 5.\" 6.\" This file contains Original Code and/or Modifications of Original Code 7.\" as defined in and that are subject to the Apple Public Source License 8.\" Version 2.0 (the 'License'). You may not use this file except in 9.\" compliance with the License. The rights granted to you under the License 10.\" may not be used to create, or enable the creation or redistribution of, 11.\" unlawful or unlicensed copies of an Apple operating system, or to 12.\" circumvent, violate, or enable the circumvention or violation of, any 13.\" terms of an Apple operating system software license agreement. 14.\" 15.\" Please obtain a copy of the License at 16.\" http://www.opensource.apple.com/apsl/ and read it before using this file. 17.\" 18.\" The Original Code and all software distributed under the License are 19.\" distributed on an 'AS IS' basis, WITHOUT WARRANTY OF ANY KIND, EITHER 20.\" EXPRESS OR IMPLIED, AND APPLE HEREBY DISCLAIMS ALL SUCH WARRANTIES, 21.\" INCLUDING WITHOUT LIMITATION, ANY WARRANTIES OF MERCHANTABILITY, 22.\" FITNESS FOR A PARTICULAR PURPOSE, QUIET ENJOYMENT OR NON-INFRINGEMENT. 23.\" Please see the License for the specific language governing rights and 24.\" limitations under the License. 25.\" 26.\" @APPLE_OSREFERENCE_LICENSE_HEADER_END@ 27.\" 28.\" @(#)posix_spawn.2 29. 30.Dd November 2, 2010 31.Dt POSIX_SPAWN 2 32.Os "Mac OS X" 33.Sh NAME 34.Nm posix_spawn 35.Nm posix_spawnp 36.Nd spawn a process 37.Sh SYNOPSIS 38.Fd #include <spawn.h> 39.Ft int 40.Fo posix_spawn 41.Fa "pid_t *restrict pid" 42.Fa "const char *restrict path" 43.Fa "const posix_spawn_file_actions_t *file_actions" 44.Fa "const posix_spawnattr_t *restrict attrp" 45.Fa "char *const argv[restrict]" 46.Fa "char *const envp[restrict]" 47.Fc 48.Ft int 49.Fo posix_spawnp 50.Fa "pid_t *restrict pid" 51.Fa "const char *restrict file" 52.Fa "const posix_spawn_file_actions_t *file_actions" 53.Fa "const posix_spawnattr_t *restrict attrp" 54.Fa "char *const argv[restrict]" 55.Fa "char *const envp[restrict]" 56.Fc 57.Sh DESCRIPTION 58The 59.Fn posix_spawn 60function creates a new process from the executable file, called the 61.Em new process file , 62specified by 63.Em path , 64which is an absolute or relative path to the file. 65The 66.Fn posix_spawnp 67function is identical to the 68.Fn posix_spawn 69function if the 70.Em file 71specified contains a slash character; otherwise, the 72.Em file 73parameter is used to construct a pathname, with its path prefix being 74obtained by a search of the path specified in the environment by the 75.Dq Ev PATH variable . 76If this variable isn't specified, the default path is set according 77to the 78.Dv _PATH_DEFPATH 79definition in 80.In paths.h , 81which is set to 82.Dq Ev /usr/bin:/bin . 83This pathname either refers to an executable object file, 84or a file of data for an interpreter; 85.Xr execve 2 86for more details. 87.Pp 88The argument 89.Fa pid 90is a pointer to a pid_t variable to receive the pid of the spawned 91process; if this is NULL, then the pid of the spawned process is 92not returned. If this pointer is non-NULL, then on successful 93completion, the variable will be modified to contain the pid of the 94spawned process. The value is undefined in the case of a failure. 95.Pp 96The argument 97.Fa file_actions 98is either NULL, or it is a pointer to a file actions object that was 99initialized by a call to 100.Xr posix_spawn_file_actions_init 3 101and represents zero or more file actions. 102.Pp 103File descriptors open in the calling process image remain open in 104the new process image, except for those for which the close-on-exec 105flag is set (see 106.Xr close 2 107and 108.Xr fcntl 2 ) . 109Descriptors that remain open are unaffected by 110.Fn posix_spawn 111unless their behaviour is modified by particular spawn flags 112or a file action; see 113.Xr posix_spawnattr_setflags 3 114and 115.Xr posix_spawn_file_actions_init 3 116for additional information. 117.Pp 118The argument 119.Fa attrp 120is either NULL, or it is a pointer to an attributes object that was 121initialized by a call to 122.Xr posix_spawnattr_init 3 123and represents a set of spawn attributes to apply. If NULL, then the 124default attributes are applied; otherwise, these attributes can control 125various aspects of the spawned process, and are applied prior to the 126spawned process beginning execution; see 127.Xr posix_spawnattr_init 3 128for more information. 129.Pp 130The argument 131.Fa argv 132is a pointer to a null-terminated array of 133character pointers to null-terminated character strings. 134These strings construct the argument list to be made available to the new 135process. At least 136.Fa argv[0] 137must be present in the array, and should contain the file name of the 138program being spawned, e.g. the last component of the 139.Em path 140or 141.Em file 142argument. 143.Pp 144The argument 145.Fa envp 146is a pointer to a null-terminated array of character pointers to 147null-terminated strings. A pointer to this array is normally stored 148in the global variable 149.Va environ. 150These strings pass information to the 151new process that is not directly an argument to the command (see 152.Xr environ 7 ) . 153.Pp 154Signals set to be ignored in the calling process are set to be ignored in 155the new process, unless the behaviour is modified by user specified 156.Em spawn attributes . 157Signals which are set to be caught in the calling process image are set to 158default action in the new process image. 159Blocked signals remain blocked regardless of changes to the signal action, 160unless the mask is overridden by user specified 161.Em spawn attributes . 162The signal stack is reset to be undefined (see 163.Xr sigaction 2 164for more information). 165.Pp 166By default, the effective user ID and group ID will be the same as those of 167the calling process image; however, this may be overridden to force them to 168be the real user ID and group ID of the parent process by user specified 169.Em spawn attributes 170(see 171.Xr posix_spawnattr_init 3 172for more information). 173.Pp 174If the set-user-ID mode bit of the new process image file is set 175(see 176.Xr chmod 2 ) , 177the effective user ID of the new process image is set to the owner ID 178of the new process image file. 179If the set-group-ID mode bit of the new process image file is set, 180the effective group ID of the new process image is set to the group ID 181of the new process image file. 182(The effective group ID is the first element of the group list.) 183The real user ID, real group ID and supplementary group IDs of the new 184process image remain the same as the calling process image. 185After any set-user-ID and set-group-ID processing, 186the effective user ID is recorded as the saved set-user-ID, 187and the effective group ID is recorded as the saved set-group-ID. 188These values may be used in changing the effective IDs later (see 189.Xr setuid 2 ) . 190.Pp 191The new process also inherits the following attributes from 192the calling process: 193.Pp 194.Bl -column parent_process_ID -offset indent -compact 195.It parent process ID Ta see Xr getppid 2 196.It process group ID Ta see Xr getpgrp 2 , Xr posix_spawnattr_init 3 197.It access groups Ta see Xr getgroups 2 198.It working directory Ta see Xr chdir 2 199.It root directory Ta see Xr chroot 2 200.It control terminal Ta see Xr termios 4 201.It resource usages Ta see Xr getrusage 2 202.It interval timers Ta see Xr getitimer 2 203.It resource limits Ta see Xr getrlimit 2 204.It file mode mask Ta see Xr umask 2 205.It signal mask Ta see Xr sigaction 2 , Xr sigsetmask 2 , 206.Xr posix_spawnattr_init 3 207.El 208.Pp 209When a program is executed as a result of a 210.Fn posix_spawn 211or 212.Fn posix_spawnp 213call, it is entered as follows: 214.Bd -literal -offset indent 215.Ft int 216.Fn main "int argc" "char **argv" "char **envp"; 217.Ed 218.Pp 219where 220.Fa argc 221is the number of elements in 222.Fa argv 223(the ``arg count'') 224and 225.Fa argv 226points to the array of character pointers 227to the arguments themselves. 228.Sh RETURN VALUES 229If the 230.Em pid 231argument is NULL, no pid is returned to the calling process; if it is 232non-NULL, then 233.Fn posix_spawn 234and 235.Fn posix_spawnp 236functions return the process ID of the child process into the pid_t 237variable pointed to by the 238.Em pid 239argument and return a 0 on success. If an error occurs, they return 240a non-zero error code as the function return value, and no child process 241is created. 242.Sh ERRORS 243The 244.Fn posix_spawn 245and 246.Fn posix_spawnp 247functions will fail and return to the calling process if: 248.Bl -tag -width Er 249.\" ========== 250.It Bq Er EINVAL 251The value specified by 252.Fa file_actions 253or 254.Fa attrp 255is invalid. 256.\" ========== 257.It Bq Er E2BIG 258The number of bytes in the new process's argument list 259is larger than the system-imposed limit. 260This limit is specified by the 261.Xr sysctl 3 262MIB variable 263.Dv KERN_ARGMAX . 264.\" ========== 265.It Bq Er EACCES 266Search permission is denied for a component of the path prefix. 267.\" ========== 268.It Bq Er EACCES 269The new process file is not an ordinary file. 270.\" ========== 271.It Bq Er EACCES 272The new process file mode denies execute permission. 273.\" ========== 274.It Bq Er EACCES 275The new process file is on a filesystem mounted 276with execution disabled 277.Pf ( Dv MNT_NOEXEC 278in 279.Ao Pa sys/mount.h Ac ) . 280.\" ========== 281.It Bq Er EFAULT 282The new process file is not as long as indicated by 283the size values in its header. 284.\" ========== 285.It Bq Er EFAULT 286.Fa Path , 287.Fa argv , 288or 289.Fa envp 290point 291to an illegal address. 292.\" ========== 293.It Bq Er EIO 294An I/O error occurred while reading from the file system. 295.\" ========== 296.It Bq Er ELOOP 297Too many symbolic links were encountered in translating the pathname. 298This is taken to be indicative of a looping symbolic link. 299.\" ========== 300.It Bq Er ENAMETOOLONG 301A component of a pathname exceeded 302.Dv {NAME_MAX} 303characters, or an entire path name exceeded 304.Dv {PATH_MAX} 305characters. 306.\" ========== 307.It Bq Er ENOENT 308The new process file does not exist. 309.\" ========== 310.It Bq Er ENOEXEC 311The new process file has the appropriate access 312permission, but has an unrecognized format 313(e.g., an invalid magic number in its header). 314.\" ========== 315.It Bq Er ENOMEM 316The new process requires more virtual memory than 317is allowed by the imposed maximum 318.Pq Xr getrlimit 2 . 319.\" ========== 320.It Bq Er ENOTDIR 321A component of the path prefix is not a directory. 322.\" ========== 323.It Bq Er ETXTBSY 324The new process file is a pure procedure (shared text) 325file that is currently open for writing or reading by some process. 326.\" ========== 327.It Bq Er EBADARCH 328The new process file has no architectures appropriate for the current system. 329.\" ========== 330.It Bq Er EBADF 331Bad file descriptor for one or more 332.Fa file_actions . 333.El 334.Pp 335Additionally, they may fail for any of the reasons listed in 336.Xr fork 2 or 337.Xr exec 3 . 338.Sh CAVEAT 339If a program is 340.Em setuid 341to a non-super-user, but is executed when 342the real 343.Em uid 344is ``root'', then the program has some of the powers 345of a super-user as well. 346.Sh SEE ALSO 347.Xr exit 2 , 348.Xr fork 2 , 349.Xr execl 3 , 350.Xr sysctl 3 , 351.Xr environ 7 , 352.Xr posix_spawnattr_init 3 , 353.Xr posix_spawn_file_actions_init 3 , 354.Sh STANDARDS 355.St -susv3 [SPN] 356.Sh HISTORY 357The 358.Fn posix_spawn 359and 360.Fn posix_spawnp 361function calls appeared in 362.St -susv3 [SPN] . 363