Openwall GNU/*/Linux - a small security-enhanced Linux distro for servers
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Date: Wed, 5 Sep 2012 07:41:58 +0400
From: Solar Designer <solar@...nwall.com>
To: john-dev@...ts.openwall.com
Subject: Re: memory usage (was: [john-users] JTR against 135 millions MD5 hashes)

On Tue, Sep 04, 2012 at 08:57:21AM +0200, magnum wrote:
> I do not know of *any* format using larger alignment than ARCH_WORD_32 except for salts (where we sometimes pass a pointer).

I don't see how salts are different from binaries in terms of pointer
passing.  I think we're doing that for both things.

There are some casts to (ARCH_WORD *) at least in DES_fmt.c when built
without DES_BS.  I guess there are such things elsewhere as well.
You're right that there are probably few of them, so we might be able to
reduce DEFAULT_ALIGN to 4 after specifying explicit alignments in those
few formats.

> BTW nearly all formats introduced since we implemented this got the DEFAULT_ALIGN so the problem has got worse instead of better.

It only got worse compared to older bleeding, where we briefly had
alignment autodetection based on allocation size.  That code was not in
any released version, nor in magnum-jumbo, so the problem hasn't gotten
worse compared to those.

Alexander

Powered by blists - more mailing lists

Your e-mail address:

Powered by Openwall GNU/*/Linux - Powered by OpenVZ