man2/setfsuid.2

   1 .\" Copyright (C) 1995, Thomas K. Dyas <tdyas@eden.rutgers.edu>
   2 .\"
   3 .\" %%%LICENSE_START(VERBATIM)
   4 .\" Permission is granted to make and distribute verbatim copies of this
   5 .\" manual provided the copyright notice and this permission notice are
   6 .\" preserved on all copies.
   7 .\"
   8 .\" Permission is granted to copy and distribute modified versions of this
   9 .\" manual under the conditions for verbatim copying, provided that the
  10 .\" entire resulting derived work is distributed under the terms of a
  11 .\" permission notice identical to this one.
  12 .\"
  13 .\" Since the Linux kernel and libraries are constantly changing, this
  14 .\" manual page may be incorrect or out-of-date.  The author(s) assume no
  15 .\" responsibility for errors or omissions, or for damages resulting from
  16 .\" the use of the information contained herein.  The author(s) may not
  17 .\" have taken the same level of care in the production of this manual,
  18 .\" which is licensed free of charge, as they might when working
  19 .\" professionally.
  20 .\"
  21 .\" Formatted or processed versions of this manual, if unaccompanied by
  22 .\" the source, must acknowledge the copyright and authors of this work.
  23 .\" %%%LICENSE_END
  24 .\"
  25 .\" Created   1995-08-06 Thomas K. Dyas <tdyas@eden.rutgers.edu>
  26 .\" Modified  2000-07-01 aeb
  27 .\" Modified  2002-07-23 aeb
  28 .\" Modified, 27 May 2004, Michael Kerrisk <mtk.manpages@gmail.com>
  29 .\"     Added notes on capability requirements
  30 .\"
  31 .TH SETFSUID 2 2016-07-17 "Linux" "Linux Programmer's Manual"
  32 .SH NAME
  33 setfsuid \- set user identity used for filesystem checks
  34 .SH SYNOPSIS
  35 .B #include <sys/fsuid.h>
  36 .PP
  37 .BI "int setfsuid(uid_t " fsuid );
  38 .SH DESCRIPTION
  39 The system call
  40 .BR setfsuid ()
  41 changes the value of the caller's filesystem user ID\(emthe
  42 user ID that the Linux kernel uses to check for all accesses
  43 to the filesystem.
  44 Normally, the value of
  45 the filesystem user ID
  46 will shadow the value of the effective user ID.
  47 In fact, whenever the
  48 effective user ID is changed,
  49 the filesystem user ID
  50 will also be changed to the new value of the effective user ID.
  51 .PP
  52 Explicit calls to
  53 .BR setfsuid ()
  54 and
  55 .BR setfsgid (2)
  56 are usually used only by programs such as the Linux NFS server that
  57 need to change what user and group ID is used for file access without a
  58 corresponding change in the real and effective user and group IDs.
  59 A change in the normal user IDs for a program such as the NFS server
  60 is a security hole that can expose it to unwanted signals.
  61 (But see below.)
  62 .PP
  63 .BR setfsuid ()
  64 will succeed only if the caller is the superuser or if
  65 .I fsuid
  66 matches either the caller's real user ID, effective user ID,
  67 saved set-user-ID, or current filesystem user ID.
  68 .SH RETURN VALUE
  69 On both success and failure,
  70 this call returns the previous filesystem user ID of the caller.
  71 .SH VERSIONS
  72 This system call is present in Linux since version 1.2.
  73 .\" This system call is present since Linux 1.1.44
  74 .\" and in libc since libc 4.7.6.
  75 .SH CONFORMING TO
  76 .BR setfsuid ()
  77 is Linux-specific and should not be used in programs intended
  78 to be portable.
  79 .SH NOTES
  80 At the time when this system call was introduced, one process
  81 could send a signal to another process with the same effective user ID.
  82 This meant that if a privileged process changed its effective user ID
  83 for the purpose of file permission checking,
  84 then it could become vulnerable to receiving signals
  85 sent by another (unprivileged) process with the same user ID.
  86 The filesystem user ID attribute was thus added to allow a process to
  87 change its user ID for the purposes of file permission checking without
  88 at the same time becoming vulnerable to receiving unwanted signals.
  89 Since Linux 2.0, signal permission handling is different (see
  90 .BR kill (2)),
  91 with the result that a process change can change its effective user ID
  92 without being vulnerable to receiving signals from unwanted processes.
  93 Thus,
  94 .BR setfsuid ()
  95 is nowadays unneeded and should be avoided in new applications
  96 (likewise for
  97 .BR setfsgid (2)).
  98 .PP
  99 The original Linux
 100 .BR setfsuid ()
 101 system call supported only 16-bit user IDs.
 102 Subsequently, Linux 2.4 added
 103 .BR setfsuid32 ()
 104 supporting 32-bit IDs.
 105 The glibc
 106 .BR setfsuid ()
 107 wrapper function transparently deals with the variation across kernel versions.
 108 .SS C library/kernel differences
 109 In glibc 2.15 and earlier,
 110 when the wrapper for this system call determines that the argument can't be
 111 passed to the kernel without integer truncation (because the kernel
 112 is old and does not support 32-bit user IDs),
 113 they will return \-1 and set \fIerrno\fP to
 114 .B EINVAL
 115 without attempting
 116 the system call.
 117 .SH BUGS
 118 No error indications of any kind are returned to the caller,
 119 and the fact that both successful and unsuccessful calls return
 120 the same value makes it impossible to directly determine
 121 whether the call succeeded or failed.
 122 Instead, the caller must resort to looking at the return value
 123 from a further call such as
 124 .IR setfsuid(\-1)
 125 (which will always fail), in order to determine if a preceding call to
 126 .BR setfsuid ()
 127 changed the filesystem user ID.
 128 At the very
 129 least,
 130 .B EPERM
 131 should be returned when the call fails (because the caller lacks the
 132 .B CAP_SETUID
 133 capability).
 134 .SH SEE ALSO
 135 .BR kill (2),
 136 .BR setfsgid (2),
 137 .BR capabilities (7),
 138 .BR credentials (7)