public inbox for glibc-bugs@sourceware.org
help / color / mirror / Atom feed
From: "jgeisler at cse dot taylor dot edu" <sourceware-bugzilla@sourceware.org>
To: glibc-bugs@sources.redhat.com
Subject: [Bug libc/11397] New: calls to cuserid() can result in buffer overflows
Date: Thu, 18 Mar 2010 03:25:00 -0000	[thread overview]
Message-ID: <20100318032449.11397.jgeisler@cse.taylor.edu> (raw)

Since cuserid() uses __getpwuid_r() to obtain the user name and then strncpy()
to copy the user name to the final buffer, the buffer could end up with a
non-NUL terminated value if the login name length >= L_cuserid.  This seems
wrong since L_cuserid is defined as 9 (8 default characters + 1 NUL) for the old
setting when all usernames never exceeded 8 characters.  It seems to me that
since strncpy() is being used, we don't worry about maintaining the final result
when it is too long to fit in a buffer of L_cuserid characters.  Therefore, we
should be safe and free to cut off one additional character to ensure the buffer
truly is NUL-terminated.

As far as I can determine, this bug has existed in glibc for an extremely long
time since this interface is deprecated and probably not actively examined for
this kind of issue.  I only found it due to a legacy app that was crashing on me
when I was logged in with long usernames, but not short usernames.

My suggested patch for sysdeps/posix/cuserid.c is as follows:

--- cuserid.c   2001-12-14 02:00:48.000000000 -0500
+++ cuserid.c.new       2010-03-17 23:08:38.000000000 -0400
@@ -44,5 +44,6 @@
 
   if (s == NULL)
     s = name;
-  return strncpy (s, pwptr->pw_name, L_cuserid);
+  s[L_cuserid - 1] = '\0';
+  return strncpy (s, pwptr->pw_name, L_cuserid - 1);
 }

-- 
           Summary: calls to cuserid() can result in buffer overflows
           Product: glibc
           Version: unspecified
            Status: NEW
          Severity: critical
          Priority: P2
         Component: libc
        AssignedTo: drepper at redhat dot com
        ReportedBy: jgeisler at cse dot taylor dot edu
                CC: glibc-bugs at sources dot redhat dot com,jgeisler at cse
                    dot taylor dot edu


http://sourceware.org/bugzilla/show_bug.cgi?id=11397

------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.


             reply	other threads:[~2010-03-18  3:25 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-03-18  3:25 jgeisler at cse dot taylor dot edu [this message]
2010-03-18 12:14 ` [Bug libc/11397] calls to cuserid() can result in buffer overruns and/or overflows jgeisler at cse dot taylor dot edu
2010-03-18 13:21 ` ldv at altlinux dot org
2010-03-18 15:55 ` jgeisler at cse dot taylor dot edu
2010-03-24 23:03 ` drepper at redhat dot com

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20100318032449.11397.jgeisler@cse.taylor.edu \
    --to=sourceware-bugzilla@sourceware.org \
    --cc=glibc-bugs@sources.redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).