Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <50FAAD1C.70109@gmail.com>
Date: Sat, 19 Jan 2013 12:26:36 -0200
From: Claudio André <claudioandre.br@...il.com>
To: john-dev@...ts.openwall.com
Subject: Re: Mulit-gpu using claudio's interfaces

Em 19-01-2013 12:19, Sayantan Datta escreveu:
>
> On Sat, Jan 19, 2013 at 7:24 PM, Claudio André 
> <claudioandre.br@...il.com <mailto:claudioandre.br@...il.com>> wrote:
>
>     while (anyGpu.hasWork_toDo) {
>
>       // ****************************************************************
>       // ***** (This, so) core can check all candidates from 16000 to
>     32000, while any other tasks inside other GPUs have not finished.
>       if (Gpu[x].finished)
>         core_send_event(result = WORK_DONE, start = 16000, finish = 32000)
>     }
>
>
> You mean core can check from 0 to 16000,while any other tasks inside 
> GPUs have not finished. Right ?
Yes (from 16000 to 32000)

>
> However ,I think myrice has done some work on generating passwords and 
> checking them on GPU itself, which seems to be a better option. Only 
> the valid candidates needs to be send back to host.
>

It is not going to solve every detail. Suppose you have a huge wordlist 
to attack on a slow hash (feasible). I prefer to have some CPU code 
working in this case.

Claudio

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.