* Andrew Makhorin <[EMAIL PROTECTED]> [2007-09-15 15:59]:
> > We are not talking about normal users following the conventions. We are
> > talking about a malicious hacker that could exploit the buffer overflow
> > vulnerability currently in GLPK. I do not know much about such exploits
> > (and h
>> > Actually, there is a way to circumvent the problem. It is quite ugly
>> > and I am not sure you would like to implement it. Here is it: make
>> > the functions xprintf and xprint1 both static, and move them,
>> > together with xvprintf, into a header file (.h) which will be
>> > included by al
* Andrew Makhorin <[EMAIL PROTECTED]> [2007-09-15 14:38]:
> > Actually, there is a way to circumvent the problem. It is quite ugly
> > and I am not sure you would like to implement it. Here is it: make
> > the functions xprintf and xprint1 both static, and move them,
> > together with xvprintf, in
> Actually, there is a way to circumvent the problem. It is quite ugly
> and I am not sure you would like to implement it. Here is it: make
> the functions xprintf and xprint1 both static, and move them,
> together with xvprintf, into a header file (.h) which will be
> included by all other files c
* Rafael Laboissiere <[EMAIL PROTECTED]> [2007-09-14 20:50]:
> * Andrew Makhorin <[EMAIL PROTECTED]> [2007-09-14 15:04]:
>
> > > Even though _glp_lib_xprintf is not declared in glpk.h, it is available in
> > > libglpk.so and malicious programs *_can_* be written that could exploit
> > > the
> >
* Andrew Makhorin <[EMAIL PROTECTED]> [2007-09-14 15:04]:
> > Even though _glp_lib_xprintf is not declared in glpk.h, it is available in
> > libglpk.so and malicious programs *_can_* be written that could exploit the
> > vulnerability.
>
> I see no way how to hide such internal routines from
>> _glp_lib_xprintf is *not* api routine and formally being not declared
>> in glpk.h it is not available to the user.
> Unfortunately, this is not the case. The following works here with GLPK
> 4.21:
> $ cat test.c
> main () { _glp_lib_xprintf ("Ouch!\n"); }
> $ gcc test.c -o test -
* Andrew Makhorin <[EMAIL PROTECTED]> [2007-09-14 13:13]:
> Friday, September 14, 2007, 12:21:09 PM, you wrote:
>
> > I am a bit confused here: xvprintf is called by xprintf in
> > src/glplib05.c. The xprintf function is actually available in the
> > public API through _glp_lib_xprintf. It would t
Friday, September 14, 2007, 12:21:09 PM, you wrote:
> * Andrew Makhorin <[EMAIL PROTECTED]> [2007-09-11 10:32]:
>> > static void
>> > xvprintf (const char *fmt, va_list arg)
>> > {
>> > charbuf[4000 + 1];
>> > vsprintf (buf, fmt, arg);
>> > xassert (strlen (buf) < size
* Andrew Makhorin <[EMAIL PROTECTED]> [2007-09-11 10:32]:
> > static void
> > xvprintf (const char *fmt, va_list arg)
> > {
> > charbuf[4000 + 1];
> > vsprintf (buf, fmt, arg);
> > xassert (strlen (buf) < sizeof (buf)); /* here! */
> > xputs (buf);
> >
>> As I remember the previous bug report concerned one informational
>> message (from glp_simplex) not masked by msg_lev = GLP_MSG_OFF, and
>> that bug was fixed in 4.21.
> Hi, yes, indeed, and thank you.
> Well, speaking tangentially I'm not sure I'd call it "fixed" to be
> absolutely precise, r
"Also sprach Andrew Makhorin:"
> As I remember the previous bug report concerned one informational
> message (from glp_simplex) not masked by msg_lev = GLP_MSG_OFF, and
> that bug was fixed in 4.21.
Hi, yes, indeed, and thank you.
Well, speaking tangentially I'm not sure I'd call it "fixed" to be
masked by msg_lev = GLP_MSG_OFF, and
that bug was fixed in 4.21.
> - Forwarded message from "Peter T. Breuer" <[EMAIL PROTECTED]> -
> From: "Peter T. Breuer" <[EMAIL PROTECTED]>
> Subject: [Pkg-scicomp-devel] Bug#441478: libglpk0: security flaw buf
Package: libglpk0
Version: 4.20-1
Severity: minor
Looking through the code for a way to shut off the annoying messaged
from lpx_adv_basis in the version of libglpk0 that is current for my
distribution (a new version has been made available to unstable, but
that's not me ...), I noticed the follow
14 matches
Mail list logo
