Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Date: Tue, 14 Nov 2017 19:17:32 +0100
From: Solar Designer <solar@...nwall.com>
To: john-dev@...ts.openwall.com
Subject: Re: Why would crypt_all() need/want to generate additional candidate passwords?

On Tue, Nov 14, 2017 at 07:04:32PM +0100, Frank Dittrich wrote:
> Why would crypt_all() generate additional candidate passwords on its own?
> Are there any sample formats which make use of this weird feature?

This was one of several formats interface changes I introduced in April
2013, and we're making heavy use of this feature for on-device (GPU and
FPGA) mask mode support (including along with other modes).

I also envisioned making use of this feature for faster LM hash support
on CPU, where crypt_all() would alter already-transposed bitslice DES
key buffers according to a mask (or maybe in an LM hash specialized
simpler exhaustive search mode), but we never cared to implement that.

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.