From: lcottrell@popmail.ucsd.edu (Lance Cottrell)
Date: Mon, 5 Dec 94 13:11:05 PST
To: cypherpunks@toad.com
Subject: Re: PGP Tools under Linux (Pr0duct Cypher)
Message-ID: <199412052110.NAA24226@ucsd.edu>
MIME-Version: 1.0
Content-Type: text/plain


-----BEGIN PGP SIGNED MESSAGE-----

To: Pr0duct Cypher

Pr0duct Cypher writes:
>I wrote:
>>I am using PGPTools to handle all the crypto functions in the second
>>generation remailer that I have written. All the copies of PGPTools that I
>>have found seem to be set up to compile on SUN workstations only.
>
>That's interesting, because I developed it under Linux and MS/DOS, and there
>was a lot of trouble at the beginning with it not running on big-endian
>machines. It sounds like whoever "fixed" it for sparcs broke it for
>little-endian machines. The original should still be on ftp.csn.net; get it
>there and it should work.
>
>>Mixmaster is both a remailer and a front end, and therefor must run on many
>>different platforms. I have been able to compile PGPTools on Linux and
>>FreeBSD, but PGPTools produces invalid RSA blocks. If you (or anyone else
>>reading this) have ported PGPTools to other platforms could you send me the
>>source and makefile, or point me to them. If not, could you help me do it,
>>the source code is a tangle, and it will take me a long time to do on my own
>>(I am a brute force programer, not a C wizard).
>
>Get the original, test it, and post the results.
>
<SNIP>
>                                                  - Pr0duct Cypher

Ok, I grabbed second oldest version (this included a bugfix).
I compiles and seems to work fine when I use it with PTD.
It fails when run under Mixmaster. The failure is deep inside pgptools.
When I call pgpk_findkey to get a key from a public key ring (by name not
number), it chugs along checking each key id in the ring. Each time it
checks a key, it uses a pair of fifo's which it then destroys. After checking
the second id, when it is destroying kc (k is destroyed without problems both
times), I get a segmentation violation in the line
   nfb=fb->next;
Since kc was simply a copy of k
  kc=fifo_copy(k);
I suspect the problem lies in fifo_copy. These are all mem fifo's.
If you want to look at the source code I am playing with (with tons of debugging
printfs in it) let me know and I will stick it on my ftp.
-----BEGIN PGP SIGNATURE-----
Version: 2.6

iQCVAwUBLuOBVFVkk3dax7hlAQGYNAP9GvDMvdSKA1vUsmJB/xitPVsiazNeZjXJ
NrjwouMg85VR1ri1rm1y2NuKbHEljGfwNB7z8euy7e1clptRsN4WWrP+ASSfrire
9Clan80Ep4lEA+83caArQv3Hcdb8+8FnLpXd/zrPI2V/3TVM2OrfU7/RtUo+3Pr3
JE2d0s6Jn+E=
=jlbs
-----END PGP SIGNATURE-----

-- 
Lance Cottrell     <lcottrell@popmail.ucsd.edu>
http://nately.ucsd.edu/~loki/
Home of the remailer chaining script "chain".
PGP 2.6 key available by finger or server.

"Love is a snowmobile racing across the tundra.  Suddenly
it flips over, pinning you underneath.  At night the ice
weasels come."
                        --Nietzsche