[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Sat, 18 Aug 2012 20:37:25 +0400
From: Solar Designer <solar@...nwall.com>
To: john-dev@...ts.openwall.com
Cc: Vladimir Vorontsov <vladimir.vorontsov@...ec.ru>
Subject: salted raw MD5 on GPU (was: [john-users] Salted MD5)
myrice, Jim, Vladimir -
On Sun, Aug 19, 2012 at 12:26:12AM +0800, myrice wrote:
> Do we implement 2bytes salt and 8bytes as separate format?
Implement variable salt length, like what dynamic_4 supports (I think).
Not just 2 and 8, but a range of lengths.
Jim - do you think dynamic_4 would work for Vladimir's needs (except
that he wants this on GPU, indeed)? Does OSC use ASCII or binary
salts? Does this require hex-encoding the salt?
> What will the format ciphertext like? If we have two salt length in
> one, I think we could:
>
> "$SaltMD5o$123456$c02e8eef3eaa1a813c2ff87c1780f9ed","test1"
> and
> "$SaltMD5b$123456$c02e8eef3eaa1a813c2ff87c1780f9ed","test1"
The above looks weird to me: both examples actually specify the same
salt (meaning same length too).
Can you just support the dynamic_4 syntax, but implement it on GPU?
As an alternative, support a syntax similar to that used by
md4_gen_fmt_plug.c, but for MD5 and on GPU. It uses:
static struct fmt_tests tests[] = {
{"$MD4p$salt$15ad2b7a23e5088942f9d3772181b384", "password"},
{"$MD4s$salt$fb483dbef17c51c13e2322fcbec5da79", "password"},
{NULL}
};
where "p" is for prefix (what Vladimir needs), "s" is for suffix (not
needed now). What you'd implement would be:
$MD5p$salt$hash
Of course, it'd only work for ASCII salts. I don't know if Vladimir
needs binary (hex-encoded) salts as well or not.
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.
