Date: Tue, 26 Nov 2013 13:02:07 +1100 From: Mark Andrews <marka@isc.org> To: Michael Butler <imb@protected-networks.net> Cc: freebsd-stable@freebsd.org Subject: Re: ipfw table add problem Message-ID: <20131126020208.37AF9AD4D2C@rock.dv.isc.org> In-Reply-To: Your message of "Mon, 25 Nov 2013 20:30:57 -0500." <5293F9D1.2030800@protected-networks.net> References: <CAAcX-AGDZbFn5RmhLBBn2PPWRPcsFUnea5MgTc7nuXGD8Ge53A@mail.gmail.com> <52911993.8010108@ipfw.ru> <CAAcX-AEt_i8RUfmMy6WLnER0X=uLk5A1=oj911k-nyMJEghRLw@mail.gmail.com> <529259DE.2040701@FreeBSD.org> <20131125152238.S78756@sola.nimnet.asn.au> <1385391778.1220.4.camel@revolution.hippie.lan> <20131126001806.27951AD3DBF@rock.dv.isc.org> <5293EBD6.8010009@protected-networks.net> <20131126003941.1ACD9AD41B4@rock.dv.isc.org> <5293F9D1.2030800@protected-networks.net>
next in thread | previous in thread | raw e-mail | index | archive | help
I remember incorrectly, the actual change was this one.
revision 1.8
date: 2001-07-16 13:22:24 +1000; author: marka; state: Exp; lines: +4 -2;
1242. [bug] inet_pton() failed to reject octal input.
Index: inet/inet_pton.c
===================================================================
RCS file: /proj/cvs/prod/bind8/src/lib/inet/inet_pton.c,v
retrieving revision 1.7
retrieving revision 1.8
diff -u -r1.7 -r1.8
--- inet/inet_pton.c 13 Oct 1999 16:39:28 -0000 1.7
+++ inet/inet_pton.c 16 Jul 2001 03:22:24 -0000 1.8
@@ -16,7 +16,7 @@
*/
#if defined(LIBC_SCCS) && !defined(lint)
-static const char rcsid[] = "$Id: inet_pton.c,v 1.7 1999-10-13 16:39:28 vixie Exp $";
+static const char rcsid[] = "$Id: inet_pton.c,v 1.8 2001-07-16 03:22:24 marka Exp $";
#endif /* LIBC_SCCS and not lint */
#include "port_before.h"
@@ -95,10 +95,12 @@
if ((pch = strchr(digits, ch)) != NULL) {
u_int new = *tp * 10 + (pch - digits);
+ if (saw_digit && *tp == 0)
+ return (0);
if (new > 255)
return (0);
*tp = new;
- if (! saw_digit) {
+ if (!saw_digit) {
if (++octets > 4)
return (0);
saw_digit = 1;
Now if you want to make it accept 0+[89][0-9]* go ahead but
0+[01234567][01234567]* need to be rejected when the inet_aton
accepts octal and hexadecimal. Having different routines return
different values for 070.070.070.070 is infintitely worse than
having 070.070.070.070 be rejected by one. Yes there is a small
subset of 8 octal inputs that match decimal input for which there
is no harm. I'm much more worried about the ones that do differ.
Mark
In message <5293F9D1.2030800@protected-networks.net>, Michael Butler writes:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> On 11/25/13 19:39, Mark Andrews wrote:
>
> >> When inet_pton() implementations have been rejecting leading zero's since
> >> they were first written their can't be a POLA. There can be a difference
> >> but not a POLA. To make is now accept a leading zero would be a POLA as
> >> there is code that depends on leading zeros being rejected by it.
>
> History disagrees with you; from the BIND 4 sources:
>
> /* This is from the BIND 4.9.4 release, modified to compile by itself */
>
> /* Copyright (c) 1996 by Internet Software Consortium.
> *
> * 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 INTERNET SOFTWARE CONSORTIUM
> DISCLAIMS
> * ALL WARRANTIES WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED
> WARRANTIES
> * OF MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL INTERNET SOFTWARE
> * CONSORTIUM 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.
> */
>
> #if defined(LIBC_SCCS) && !defined(lint)
> static char rcsid[] = "$Id: inet_pton.c,v 8.6 1996/06/26 23:17:26 vixie
> Exp $";
> #endif /* LIBC_SCCS and not lint */
>
> [ .. snip .. ]
>
> /* int
> * inet_pton4(src, dst)
> * like inet_aton() but without all the hexadecimal and shorthand.
> * return:
> * 1 if `src' is a valid dotted quad, else 0.
> * notice:
> * does not touch `dst' unless it's returning 1.
> * author:
> * Paul Vixie, 1996.
> */
> static int
> inet_pton4(src, dst)
> const char *src;
> u_char *dst;
> {
> static const char digits[] = "0123456789";
> int saw_digit, octets, ch;
> u_char tmp[INADDRSZ], *tp;
>
> saw_digit = 0;
> octets = 0;
> *(tp = tmp) = 0;
> while ((ch = *src++) != '\0') {
> const char *pch;
>
> if ((pch = strchr(digits, ch)) != NULL) {
> u_int new = *tp * 10 + (pch - digits);
>
> if (new > 255)
> return (0);
> *tp = new;
> if (! saw_digit) {
> if (++octets > 4)
> return (0);
> saw_digit = 1;
> }
> } else if (ch == '.' && saw_digit) {
> if (octets == 4)
> return (0);
> *++tp = 0;
> saw_digit = 0;
> } else
> return (0);
> }
> if (octets < 4)
> return (0);
> /* bcopy(tmp, dst, INADDRSZ); */
> memcpy(dst, tmp, INADDRSZ);
> return (1);
> }
>
> Note especially that at some time after 1996, an additional two lines
> were added and the man page not updated to reflect their addition.
>
> Above the test "(new > 255)" these lines were added:
>
> if (saw_digit && *tp == 0)
> return (0);
>
> .. and which now causes this confusion.
>
> Either the code or the man page are wrong,
>
> imb
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.15 (FreeBSD)
>
> iEYEARECAAYFAlKT+dEACgkQQv9rrgRC1JJOxwCgpqtkHUv+d15C49JPpnNrwPI/
> zrkAmwSXukkcAhAKKfteEuFRe7tbTqnT
> =1JzH
> -----END PGP SIGNATURE-----
--
Mark Andrews, ISC
1 Seymour St., Dundas Valley, NSW 2117, Australia
PHONE: +61 2 9871 4742 INTERNET: marka@isc.org
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20131126020208.37AF9AD4D2C>