[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Thu, 30 May 2013 16:52:06 -0300
From: Claudio André <claudioandre.br@...il.com>
To: john-dev@...ts.openwall.com
Subject: RE: Unify internal form of mscash2 hashes?
Sorry, i don't have a computer and this is not a complete patch. It was an
example to show my point of view.
1. agree about cuda, but i won't change any file that is not 'mine'. I
won't be able to fix bugs inside them (and i will create new ones).
2. ROUNS-PREFIX and other details should be moved. They will.
The patch is not that bad, as is. It is not complete, of course. Magnum
commit just to put pressure on format maintainers, i guess.
I do not intend to have a dev machine anymore, but as soon as i get
something better than a tablet, i can fix these problems.
Claudio
Em 30/05/2013 16:20, "jfoug" <jfoug@....net> escreveu:
>
> From: Claudio André Sent: Thursday, May 30, 2013 13:02
> >Sorry, one file missing.
>
> Shouldn't the ROUNDS_PREFIX define be within the cryptsha512_valid.h file
also? Or is it only valid to have that string in only one of the
underlying formats?
>
> Also, the cuda version should use this same header file, and have it's
in-source valid() function removed also. The whole exercise is to get down
to only ONE instance of that valid() that is shared by all of the same
formats which may have implementations for various architectures in
different files, so that these pre-filtration functions (which are NOT part
of the core architecture specific code), is exactly the same for all of
them.
>
> Jim.
>
Content of type "text/html" skipped
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.
