Mailing List Archive

This is a patch to avoid the redefinition of the uintxx_t structures.

Ben

--- defines.h.ORIG Thu Dec 30 23:05:41 1999
+++ defines.h Thu Dec 30 23:06:35 1999
@@ -61,7 +61,7 @@
# else
# if (SIZEOF_LONG_LONG_INT == 8)
typedef long long int int64_t;
-# define HAVE_INTXX_T
+# define HAVE_INTXX_T 1
# else
# error "64 bit int type not found."
# endif
@@ -74,6 +74,7 @@
typedef uint16_t u_int16_t;
typedef uint32_t u_int32_t;
typedef uint64_t u_int64_t;
+# define HAVE_U_INTXX_T 1
# else
# if (SIZEOF_SHORT_INT == 2)
typedef unsigned short int u_int16_t;

Upgraded to this one, got it compiled and installed, and now I'm getting:

> ssh dragon
Local: Corrupted check bytes on input.

dragon:/var/src/openssh-1.2.1pre24> /usr/slocal/sbin/sshd -d
debug: sshd version OpenSSH-1.2.1
Server listening on port 22.
Generating 768 bit RSA key.
RSA key generation complete.
debug: Server will not fork when running in debugging mode.
Connection from 131.162.138.223 port 636
debug: Client protocol version 1.5; client software version 1.2.26
debug: Sent 768 bit public key and 1024 bit host key.
debug: Encryption type: 3des
debug: Received session key; encryption turned on.
debug: Installing crc compensation attack detector.
Disconnecting: Corrupted check bytes on input.
debug: Calling cleanup 0x806028c(0x0)

Tried regenerating the server key, and removing it from my
.ssh/known_hosts and regenerating there...no change...

Not sure where else to look ...

On Fri, 31 Dec 1999, Damien Miller wrote:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> openssh-1.2.1pre24 is being uploaded to:
>
> http://violet.ibs.com.au/openssh/files/
>
> This release fixes the silly bugs (almost all autoconf related) that
> crept into yesterday's release.
>
> 19991231
> - Fix password support on systems with a mixture of shadowed and
> non-shadowed passwords (e.g. NIS). Report and fix from
> HARUYAMA Seigo <haruyama@nt.phys.s.u-tokyo.ac.jp>
> - Fix broken autoconf typedef detection. Report from Marc G.
> Fournier <marc.fournier@acadiau.ca>
> - Fix occasional crash on LinuxPPC. Patch from Franz Sirl
> <Franz.Sirl-kernel@lauterbach.com>
> - Prevent typedefs from being compiled more than once. Report from
> Marc G. Fournier <marc.fournier@acadiau.ca>
> - Fill in ut_utaddr utmp field. Report from Benjamin Charron
> <iretd@bigfoot.com>
> - Really fix broken default path. Fix from Jim Knoble
> <jmknoble@pobox.com>
> - Remove test for quad_t. No longer needed.
>
> Regards,
> Damien
>
> - --
> | "Bombay is 250ms from New York in the new world order" - Alan Cox
> | Damien Miller - http://www.mindrot.org/
> | Email: djm@mindrot.org (home) -or- djm@ibs.com.au (work)
>
>
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.0.0 (GNU/Linux)
> Comment: For info see http://www.gnupg.org
>
> iD8DBQE4a+7dormJ9RG1dI8RAtPVAJ9eei2hVKSrLRhk5tDNjI6sIn/ybQCeNhBr
> INmewdyMfjU0SV6xlVqb34M=
> =6Bj8
> -----END PGP SIGNATURE-----
>
>
>

Marc G. Fournier marc.fournier@acadiau.ca
Senior Systems Administrator Acadia University

"These are my opinions, which are not necessarily shared by my employer"

On Fri, Dec 31, 1999 at 10:46:34AM +1100, Damien Miller wrote:
>
> openssh-1.2.1pre24 is being uploaded to:
>

HPUX status:
- compile problem, patch follows.
- utmp is now registering logins on the ptys (well done), but not
logouts. I'll look into it.
- manpages still broken, I hope to have a fix for that (probably
precompile them :-( ) today
- fixpaths is run twice, once at the end of a 'make', and again for a
'make install'

HPUX croaks, as I suspect will other plats, at a multi-include of
defines.h . Fair enough, I suppose. A patch follows, though I doubt it's
necessary in this case :-) Here's the trace:

includes.h
...
bsd-misc.h
config.h
defines.h
<other>.h
config.h
defines.h -> error.
...
config.h
defines.h -> error.

The patch fixes the current compile problem. IMO other includes,
particularly config.h, should have guards as well. An equally good fix
may be to stop .h files including config.h, as the .c should probably do
that.

-Andre


--- openssh-1.2.1pre24.orig/defines.h Thu Dec 30 22:55:34 1999
+++ openssh-1.2.1pre24.new/defines.h Fri Dec 31 14:35:12 1999
@@ -1,3 +1,6 @@
+#ifndef __DEFINES_H__
+#define __DEFINES_H__
+
/* Necessary headers */

#include <sys/types.h> /* For u_intXX_t */
@@ -223,4 +226,6 @@
#else
# define PAM_STRERROR(a,b) pam_strerror((a),(b))
#endif
+
+#endif /* __DEFINES_H__ */


--
Andre Lucas <andre.lucas@dial.pipex.com>
http://dspace.dial.pipex.com/andre.lucas/

I've sent along another patch on this subject (same thread.) I think the
problem here is the multi-include, not the __.*int.* defines. With a
multi-include guard around defines.h, the problem goes away.

Ta,
-Andre

On Thu, Dec 30, 1999 at 11:08:24PM -0500, Ben Taylor wrote:
>
> This is a patch to avoid the redefinition of the uintxx_t structures.
>
> Ben
>
> --- defines.h.ORIG Thu Dec 30 23:05:41 1999
> +++ defines.h Thu Dec 30 23:06:35 1999
> @@ -61,7 +61,7 @@
> # else
> # if (SIZEOF_LONG_LONG_INT == 8)
> typedef long long int int64_t;
> -# define HAVE_INTXX_T
> +# define HAVE_INTXX_T 1
> # else
> # error "64 bit int type not found."
> # endif
> @@ -74,6 +74,7 @@
> typedef uint16_t u_int16_t;
> typedef uint32_t u_int32_t;
> typedef uint64_t u_int64_t;
> +# define HAVE_U_INTXX_T 1
> # else
> # if (SIZEOF_SHORT_INT == 2)
> typedef unsigned short int u_int16_t;
>

--
Andre Lucas <andre.lucas@dial.pipex.com>
http://dspace.dial.pipex.com/andre.lucas/