[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20150817192354.GA2647@openwall.com>
Date: Mon, 17 Aug 2015 22:23:55 +0300
From: Solar Designer <solar@...nwall.com>
To: john-dev@...ts.openwall.com
Subject: Re: Sybase-PROP (Re: FMT_OMP_BAD)
Dhiru -
The feal8_plug.c and syb-prop_repro_plug.c files lack proper license and
copyright statements.
feal8_plug.c starts with:
[solar@...er src]$ head -1 feal8_plug.c | iconv -cfutf8 -tascii
/* FEAL8 Implementation of NTTs FEAL-8 cipher. Version of 11 September 1989. */
A copy of the actual "Version of 11 September 1989" appears to be here:
http://cryptospecs.googlecode.com/svn/trunk/symmetrical/sources/feal8.c
and JtR jumbo's revision at:
http://www.nes.fr/securitylab/?p=1128
which also includes the syb-prop_repro_plug.c code.
I guess syb-prop_repro_plug.c is copyright'ed by the folks who
reverse-engineered the algorithm and wrote this code. (The algorithm is
not copyright'able, and the code is all newly written.) Perhaps we
simply need to add their names in there, from SybasePROP_fmt_plug.c.
And a license statement that they'd agree with (hopefully, same as in
SybasePROP_fmt_plug.c). Dhiru, can you take care of this?
As to feal8_plug.c, it appears the original lacked copyright and license
statements. Was it the reference implementation of FEAL-8? Looks like
we'd need to code our own, or find a suitably licensed one, for this to
be clean. If coding our own, we'd also include 2x or 3x interleaving
for some speedup.
While at it, please note the changes I made to handling of OMP_SCALE
(and having its value split across OMP_SCALE and MAX_KEYS_PER_CRYPT, as
tuned with some benchmarks at 1 vs. many threads). You'll want to do
similar in your other and future formats, for better performance in
non-OpenMP builds and when running just 1 thread in OpenMP builds.
Thanks,
Alexander
Powered by blists - more mailing lists
Confused about mailing lists and their use?
Read about mailing lists on Wikipedia
and check out these
guidelines on proper formatting of your messages.
