Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Sat, 26 Nov 2011 20:48:08 +0100
From: magnum <>
Subject: Re: mdfourmmx

On Sat, 26 Nov 2011, Solar wrote
>> The reason I was considering it is that it's like 3-4 extra lines of
>> code in each format (bar the #ifdefs) once the .S file is there, given
>> you already have SSExxxbody implemented in a format.
> Sorry, I don't know what you (Jim) are referring to here (the "it" in
> "I was considering it").

That block of text was written by me. Jim's mail client and mine
sometimes disagree about line wrapping so the quoting ended up a little
broken. By the way I'm *now* quoting from the mail archive, I never got
the mail I'm quoting here - or perhaps I accidentally deleted it somehow.

OT, what I meant is that it's very easy to add mdfourmmx() to a format
that already supports intrinsics.

> OK.  It sounds like the three of you (JimF, magnum, Simon) really want
> to drop these older files in favor of the intrinsics. 

No, I don't mind keeping them although I have no use of them myself.
Since they use the same buffer layout as the intrinsics, there is
practically no extra work. I already have a working patch that adds
Simons's MD4 .S as posted here (but modified a little):


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.