factotum/libressl-portable
1
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

remove issetuigid wrappers, now that all getenv calls are gone.

Browse Source 
From deraadt@ upstream:

Remove all getenv() calls, especially those wrapped by issetugid().
getenv()'s wrapped by issetugid() are safe, but issetugid() is ...
difficult to impliment on many operating systems.  By accident, a grand
experiment was run over the last year, where issetugid() returned 1 (the
safe value) on a few operating systems.  Noone noticed & complained that
certain environment variables were not working.......
This commit is contained in:
Brent Cook
2015-04-14 07:55:23 -05:00
parent fe3f7fc636
commit 3096ab0e45
9 changed files with 3 additions and 244 deletions
Download Patch File Download Diff File Expand all files Collapse all files

107
crypto/compat/issetugid_aix.c
View File

@@ -1,107 +0,0 @@
/* $OpenBSD: $ */
/*
* Copyright (c) 2015 Michael Felt <aixtools@gmail.com>
*
* Permission to use, copy, modify, and distribute this software for any
* purpose with or without fee is hereby granted, provided that the above
* copyright notice and this permission notice appear in all copies.
*
* THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
* WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
* MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
* ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
* WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
* ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
* OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
*
*/
#include <sys/id.h>
#include <sys/priv.h>
#include <stdio.h>
#include <unistd.h>
/*
* AIX does not have issetugid().
* This experimental implementation uses getpriv() and get*id().
* First, try getpriv() and check equality of pv_priv values
* When these values are equal, using get*id() including login uid.
*
*/
int issetugid(void)
{
/*
* Return fail-safe while we evaluate primitives in AIX. There does
* not yet appear to be a single atomic test to tell if privileges of
* the process changed from that of the user who is in control of the
* environment.
*/
return (1);
#define PEPRIV(a,b) a.pv_priv[b]
/*
* effective priv is what I can do now
* inherited priv is what the caller gave or could have given
* basically when inherited == 0 and effective != 0 then
* some kind of priv escalation has occurred
* when 'demoted' -- inherited != 0 but effective == 0
* there is also a change, so, will report 1 as well - to be safe
* PROBABLY there needs more study re: how RBAC subtley affects
* the priv_t values - for now, they are either zero - nothing added
* or non-zero - something added
*/
priv_t effective,inherited;
int luid;
int euid, ruid;
getpriv(PRIV_EFFECTIVE, &effective, sizeof(priv_t));
getpriv(PRIV_INHERITED, &inherited, sizeof(priv_t));
if (PEPRIV(effective,0) | PEPRIV(effective,1)) { /* have something */
if ((PEPRIV(inherited,0) | PEPRIV(inherited,1)) == 0) /* had nothing - classic u+s bit */
return (1);
} else {
/*
* effective priv elevation is NULL/NONE
* was there something and removed via setuid()?
*/
if (PEPRIV(inherited,0) | PEPRIV(inherited,1))
return (1);
}
/*
* if we get this far, then "no" differences in process priv noted
* compare the different uid
* the comparision of login id with effective says "TRUE" when different.
* this may not work as expected when using sudo for elevation
* again, looking at RBAC affects on priv may be more truthful
*
* ruid - real uid
* euid - effictive uid
* luid - login uid
*/
/*
* if these differ (not common on AIX), return changed
*/
ruid = getuid();
euid = geteuid();
if (euid != ruid)
return (1);
if (getgid() != getegid())
return (1);
/*
* luid == login id, su/sudo do not/cannot change this afaik
* perhaps this is "too strict", but same as in
* issetugid_win.c - err on the safe side for now
*/
luid = getuidx(ID_LOGIN);
if (euid != luid)
return (1);
return (0);
}

17
crypto/compat/issetugid_hpux.c
View File

@@ -1,17 +0,0 @@
#include <stdio.h>
#include <unistd.h>
#include <sys/pstat.h>
/*
* HP-UX does not have issetugid().
* Use pstat_getproc() and check PS_CHANGEDPRIV bit of pst_flag. If this call
* cannot be used, assume we must be running in a privileged environment.
*/
int issetugid(void)
{
struct pst_status buf;
if (pstat_getproc(&buf, sizeof(buf), 0, getpid()) == 1 &&
!(buf.pst_flag & PS_CHANGEDPRIV))
return 0;
return 1;
}

47
crypto/compat/issetugid_linux.c
View File

@@ -1,47 +0,0 @@
/*
* issetugid implementation for Linux
* Public domain
*/
#include <errno.h>
#include <gnu/libc-version.h>
#include <string.h>
#include <sys/types.h>
#include <unistd.h>
/*
* Linux-specific glibc 2.16+ interface for determining if a process was
* launched setuid/setgid or with additional capabilities.
*/
#ifdef HAVE_GETAUXVAL
#include <sys/auxv.h>
#endif
int issetugid(void)
{
#ifdef HAVE_GETAUXVAL
/*
* The API for glibc < 2.19 does not indicate if there is an error with
* getauxval. While it should not be the case that any 2.6 or greater
* kernel ever does not supply AT_SECURE, an emulated software environment
* might rewrite the aux vector.
*
* See https://sourceware.org/bugzilla/show_bug.cgi?id=15846
*
* Perhaps this code should just read the aux vector itself, so we have
* backward-compatibility and error handling in older glibc versions.
* info: http://lwn.net/Articles/519085/
*
*/
const char *glcv = gnu_get_libc_version();
if (strverscmp(glcv, "2.19") >= 0) {
errno = 0;
if (getauxval(AT_SECURE) == 0) {
if (errno != ENOENT) {
return 0;
}
}
}
#endif
return 1;
}

16
crypto/compat/issetugid_osx.c
View File

@@ -1,16 +0,0 @@
/*
* issetugid implementation for OS X
* Public domain
*/
#include <unistd.h>
/*
* OS X has issetugid, but it is not fork-safe as of version 10.10.
* See this Solaris report for test code that fails similarly:
* http://mcarpenter.org/blog/2013/01/15/solaris-issetugid%282%29-bug
*/
int issetugid(void)
{
return 1;
}

26
crypto/compat/issetugid_win.c
View File

@@ -1,26 +0,0 @@
/*
* issetugid implementation for Windows
* Public domain
*/
#include <unistd.h>
/*
* Windows does not have a native setuid/setgid functionality.
* A user must enter credentials each time a process elevates its
* privileges.
*
* So, in theory, this could always return 0, given what I know currently.
* However, it makes sense to stub out initially in 'safe' mode until we
* understand more (and determine if any disabled functionality is actually
* useful on Windows anyway).
*
* Future versions of this function that are made more 'open' should thoroughly
* consider the case of this code running as a privileged service with saved
* user credentials or privilege escalations by other means (e.g. the old
* RunAsEx utility.)
*/
int issetugid(void)
{
return 1;
}