[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <d1f2ee6834946925a3bf2f33c9214ac0@smtp.hushmail.com>
Date: Sun, 9 Jun 2013 22:26:34 +0200
From: magnum <john.magnum@...hmail.com>
To: john-dev@...ts.openwall.com
Subject: Re: Bleeding jumbo: more GPU formats missing "split() unifies case" flag and other differences in format flags
On 2 Jun, 2013, at 23:51 , Frank Dittrich <frank_dittrich@...mail.com> wrote:
> Beside the raw-sha1-opencl, these GPU formats seem to have the same issue:
>
> raw-sha224 00020003
> raw-sha224-cuda 00000003
>
> raw-sha256 00020003
> raw-sha256-cuda 00000003
>
> raw-sha512 00020003
> raw-sha512-cuda 00000003
> raw-sha512-opencl 00000003
Until I see an uppercase SHA2 in the wild I don't think this is needed. Although it doesn't hurt fixing.
> This is the other way round, only in one GPU implementation split()
> unifies case:
> xsha512 00000003
> xsha512-cuda 00000003
> xsha512-opencl 00000003
> xsha512-ng-opencl 00020003
Same here.
> Other differences between flags of various implementations of the same
> format:
>
> False positives possible: yes vs. no:
>
> strip-opencl 00000103
> strip 00000003
We might want to rectify this but I'm not sure in what direction. Dhiru?
> ssh-ng 00000103
> ssh 00000003
This is correct. They are two completely different implementations and ssh-ng does in fact give false positives.
magnum
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.
