man2/access.2

   1 .\" This manpage is Copyright (C) 1992 Drew Eckhardt;
   2 .\"             and Copyright (C) 1993 Michael Haardt, Ian Jackson.
   3 .\" and Copyright (C) 2004, 2006, 2007, 2014 Michael Kerrisk <mtk.manpages@gmail.com>
   4 .\"
   5 .\" %%%LICENSE_START(VERBATIM)
   6 .\" Permission is granted to make and distribute verbatim copies of this
   7 .\" manual provided the copyright notice and this permission notice are
   8 .\" preserved on all copies.
   9 .\"
  10 .\" Permission is granted to copy and distribute modified versions of this
  11 .\" manual under the conditions for verbatim copying, provided that the
  12 .\" entire resulting derived work is distributed under the terms of a
  13 .\" permission notice identical to this one.
  14 .\"
  15 .\" Since the Linux kernel and libraries are constantly changing, this
  16 .\" manual page may be incorrect or out-of-date.  The author(s) assume no
  17 .\" responsibility for errors or omissions, or for damages resulting from
  18 .\" the use of the information contained herein.  The author(s) may not
  19 .\" have taken the same level of care in the production of this manual,
  20 .\" which is licensed free of charge, as they might when working
  21 .\" professionally.
  22 .\"
  23 .\" Formatted or processed versions of this manual, if unaccompanied by
  24 .\" the source, must acknowledge the copyright and authors of this work.
  25 .\" %%%LICENSE_END
  26 .\"
  27 .\" Modified 1993-07-21 Rik Faith (faith@cs.unc.edu)
  28 .\" Modified 1994-08-21 by Michael Chastain (mec@shell.portal.com):
  29 .\"   Removed note about old kernel (pre-1.1.44) using wrong id on path.
  30 .\" Modified 1996-03-18 by Martin Schulze (joey@infodrom.north.de):
  31 .\"   Stated more clearly how it behaves with symbolic links.
  32 .\" Added correction due to Nick Duffek (nsd@bbc.com), aeb, 960426
  33 .\" Modified 1996-09-07 by Michael Haardt:
  34 .\"   Restrictions for NFS
  35 .\" Modified 1997-09-09 by Joseph S. Myers <jsm28@cam.ac.uk>
  36 .\" Modified 1998-01-13 by Michael Haardt:
  37 .\"   Using access is often insecure
  38 .\" Modified 2001-10-16 by aeb
  39 .\" Modified 2002-04-23 by Roger Luethi <rl@hellgate.ch>
  40 .\" Modified 2004-06-23 by Michael Kerrisk
  41 .\" 2007-06-10, mtk, various parts rewritten, and added BUGS section.
  42 .\"
  43 .TH ACCESS 2 2014-02-21 "Linux" "Linux Programmer's Manual"
  44 .SH NAME
  45 access, faccessat \- check user's permissions for a file
  46 .SH SYNOPSIS
  47 .nf
  48 .B #include <unistd.h>
  49 .sp
  50 .BI "int access(const char *" pathname ", int " mode );
  51
  52 .B #include <fcntl.h>           /* Definition of AT_* constants */
  53 .B #include <unistd.h>
  54 .sp
  55 .BI "int faccessat(int " dirfd ", const char *" pathname ", int " \
  56 mode ", int " flags );
  57 .fi
  58 .sp
  59 .in -4n
  60 Feature Test Macro Requirements for glibc (see
  61 .BR feature_test_macros (7)):
  62 .in
  63 .sp
  64 .BR faccessat ():
  65 .PD 0
  66 .ad l
  67 .RS 4
  68 .TP 4
  69 Since glibc 2.10:
  70 _XOPEN_SOURCE\ >=\ 700 || _POSIX_C_SOURCE\ >=\ 200809L
  71 .TP
  72 Before glibc 2.10:
  73 _ATFILE_SOURCE
  74 .RE
  75 .ad
  76 .PD
  77 .fi
  78 .SH DESCRIPTION
  79 .BR access ()
  80 checks whether the calling process can access the file
  81 .IR pathname .
  82 If
  83 .I pathname
  84 is a symbolic link, it is dereferenced.
  85
  86 The
  87 .I mode
  88 specifies the accessibility check(s) to be performed,
  89 and is either the value
  90 .BR F_OK ,
  91 .\" F_OK is defined as 0 on every system that I know of.
  92 or a mask consisting of the bitwise OR of one or more of
  93 .BR R_OK ", " W_OK ", and " X_OK .
  94 .B F_OK
  95 tests for the existence of the file.
  96 .BR R_OK ", " W_OK ", and " X_OK
  97 test whether the file exists and grants read, write, and
  98 execute permissions, respectively.
  99
 100 The check is done using the calling process's
 101 .I real
 102 UID and GID, rather than the effective IDs as is done when
 103 actually attempting an operation (e.g.,
 104 .BR open (2))
 105 on the file.
 106 This allows set-user-ID programs to
 107 easily determine the invoking user's authority.
 108
 109 If the calling process is privileged (i.e., its real UID is zero),
 110 then an
 111 .B X_OK
 112 check is successful for a regular file if execute permission
 113 is enabled for any of the file owner, group, or other.
 114 .SS faccessat ()
 115 The
 116 .BR faccessat ()
 117 system call operates in exactly the same way as
 118 .BR access (2),
 119 except for the differences described here.
 120
 121 If the pathname given in
 122 .I pathname
 123 is relative, then it is interpreted relative to the directory
 124 referred to by the file descriptor
 125 .I dirfd
 126 (rather than relative to the current working directory of
 127 the calling process, as is done by
 128 .BR access (2)
 129 for a relative pathname).
 130
 131 If
 132 .I pathname
 133 is relative and
 134 .I dirfd
 135 is the special value
 136 .BR AT_FDCWD ,
 137 then
 138 .I pathname
 139 is interpreted relative to the current working
 140 directory of the calling process (like
 141 .BR access (2)).
 142
 143 If
 144 .I pathname
 145 is absolute, then
 146 .I dirfd
 147 is ignored.
 148
 149 .I flags
 150 is constructed by ORing together zero or more of the following values:
 151 .TP
 152 .B AT_EACCESS
 153 Perform access checks using the effective user and group IDs.
 154 By default,
 155 .BR faccessat ()
 156 uses the real IDs (like
 157 .BR access (2)).
 158 .TP
 159 .B AT_SYMLINK_NOFOLLOW
 160 If
 161 .I pathname
 162 is a symbolic link, do not dereference it:
 163 instead return information about the link itself.
 164 .SH "RETURN VALUE"
 165 On success (all requested permissions granted, or
 166 .I mode
 167 is
 168 .B F_OK
 169 and the file exists), zero is returned.
 170 On error (at least one bit in
 171 .I mode
 172 asked for a permission that is denied, or
 173 .I mode
 174 is
 175 .B F_OK
 176 and the file does not exist, or some other error occurred),
 177 \-1 is returned, and
 178 .I errno
 179 is set appropriately.
 180 .SH ERRORS
 181 .BR access ()
 182 and
 183 .BR faccessat ()
 184 shall fail if:
 185 .TP
 186 .B EACCES
 187 The requested access would be denied to the file, or search permission
 188 is denied for one of the directories in the path prefix of
 189 .IR pathname .
 190 (See also
 191 .BR path_resolution (7).)
 192 .TP
 193 .B ELOOP
 194 Too many symbolic links were encountered in resolving
 195 .IR pathname .
 196 .TP
 197 .B ENAMETOOLONG
 198 .I pathname
 199 is too long.
 200 .TP
 201 .B ENOENT
 202 A component of
 203 .I pathname
 204 does not exist or is a dangling symbolic link.
 205 .TP
 206 .B ENOTDIR
 207 A component used as a directory in
 208 .I pathname
 209 is not, in fact, a directory.
 210 .TP
 211 .B EROFS
 212 Write permission was requested for a file on a read-only filesystem.
 213 .PP
 214 .BR access ()
 215 and
 216 .BR faccessat ()
 217 may fail if:
 218 .TP
 219 .B EFAULT
 220 .I pathname
 221 points outside your accessible address space.
 222 .TP
 223 .B EINVAL
 224 .I mode
 225 was incorrectly specified.
 226 .TP
 227 .B EIO
 228 An I/O error occurred.
 229 .TP
 230 .B ENOMEM
 231 Insufficient kernel memory was available.
 232 .TP
 233 .B ETXTBSY
 234 Write access was requested to an executable which is being
 235 executed.
 236 .PP
 237 The following additional errors can occur for
 238 .BR faccessat ():
 239 .TP
 240 .B EBADF
 241 .I dirfd
 242 is not a valid file descriptor.
 243 .TP
 244 .B EINVAL
 245 Invalid flag specified in
 246 .IR flags .
 247 .TP
 248 .B ENOTDIR
 249 .I pathname
 250 is relative and
 251 .I dirfd
 252 is a file descriptor referring to a file other than a directory.
 253 .SH VERSIONS
 254 .BR faccessat ()
 255 was added to Linux in kernel 2.6.16;
 256 library support was added to glibc in version 2.4.
 257 .SH "CONFORMING TO"
 258 .BR access (2):
 259 SVr4, 4.3BSD, POSIX.1-2001, POSIX.1-2008.
 260
 261 .BR faccessat (2):
 262 POSIX.1-2008.
 263 .SH NOTES
 264 .PP
 265 .BR Warning :
 266 Using these calls to check if a user is authorized to, for example,
 267 open a file before actually doing so using
 268 .BR open (2)
 269 creates a security hole, because the user might exploit the short time
 270 interval between checking and opening the file to manipulate it.
 271 .BR "For this reason, the use of this system call should be avoided" .
 272 (In the example just described,
 273 a safer alternative would be to temporarily switch the process's
 274 effective user ID to the real ID and then call
 275 .BR open (2).)
 276 .PP
 277 .BR access ()
 278 always dereferences symbolic links.
 279 If you need to check the permissions on a symbolic link, use
 280 .BR faccessat (2)
 281 with the flag
 282 .BR AT_SYMLINK_NOFOLLOW .
 283 .PP
 284 These calls return an error if any of the access types in
 285 .I mode
 286 is denied, even if some of the other access types in
 287 .I mode
 288 are permitted.
 289 .PP
 290 If the calling process has appropriate privileges (i.e., is superuser),
 291 POSIX.1-2001 permits an implementation to indicate success for an
 292 .B X_OK
 293 check even if none of the execute file permission bits are set.
 294 .\" HPU-UX 11 and Tru64 5.1 do this.
 295 Linux does not do this.
 296 .PP
 297 A file is accessible only if the permissions on each of the
 298 directories in the path prefix of
 299 .I pathname
 300 grant search (i.e., execute) access.
 301 If any directory is inaccessible, then the
 302 .BR access ()
 303 call will fail, regardless of the permissions on the file itself.
 304 .PP
 305 Only access bits are checked, not the file type or contents.
 306 Therefore, if a directory is found to be writable,
 307 it probably means that files can be created in the directory,
 308 and not that the directory can be written as a file.
 309 Similarly, a DOS file may be found to be "executable," but the
 310 .BR execve (2)
 311 call will still fail.
 312 .PP
 313 These calls
 314 may not work correctly on NFSv2 filesystems with UID mapping enabled,
 315 because UID mapping is done on the server and hidden from the client,
 316 which checks permissions.  (NFS versions 3 and higher perform the check on
 317 the server.)
 318 Similar problems can occur to FUSE mounts.
 319 .\"
 320 .\"
 321 .SS faccessat ()
 322 See
 323 .BR openat (2)
 324 for an explanation of the need for
 325 .BR faccessat ().
 326
 327 .IR Warning :
 328 .BR faccessat ()
 329 is subject to the same kinds of races as
 330 .BR access (2)
 331 and
 332 .BR euidaccess (3).
 333 .SS Glibc notes
 334 The raw system call takes only the first three arguments.
 335 The
 336 .B AT_EACCESS
 337 and
 338 .B AT_SYMLINK_NOFOLLOW
 339 flags are actually implemented within the glibc wrapper function for
 340 .BR faccessat ().
 341 If either of these flags is specified, then the wrapper function employs
 342 .BR fstatat (2)
 343 to determine access permissions.
 344 .SH BUGS
 345 In kernel 2.4 (and earlier) there is some strangeness in the handling of
 346 .B X_OK
 347 tests for superuser.
 348 If all categories of execute permission are disabled
 349 for a nondirectory file, then the only
 350 .BR access ()
 351 test that returns \-1 is when
 352 .I mode
 353 is specified as just
 354 .BR X_OK ;
 355 if
 356 .B R_OK
 357 or
 358 .B W_OK
 359 is also specified in
 360 .IR mode ,
 361 then
 362 .BR access ()
 363 returns 0 for such files.
 364 .\" This behavior appears to have been an implementation accident.
 365 Early 2.6 kernels (up to and including 2.6.3)
 366 also behaved in the same way as kernel 2.4.
 367
 368 In kernels before 2.6.20,
 369 these calls ignored the effect of the
 370 .B MS_NOEXEC
 371 flag if it was used to
 372 .BR mount (2)
 373 the underlying filesystem.
 374 Since kernel 2.6.20, the
 375 .B MS_NOEXEC
 376 is honored
 377 .SH "SEE ALSO"
 378 .BR chmod (2),
 379 .BR chown (2),
 380 .BR open (2),
 381 .BR setgid (2),
 382 .BR setuid (2),
 383 .BR stat (2),
 384 .BR euidaccess (3),
 385 .BR credentials (7),
 386 .BR path_resolution (7),
 387 .BR symlink (7)