[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CAN4B415MMJ=Bse0aBaV6FYrq1-LTv6EhYVTAAdehjAKt7GwJXw@mail.gmail.com>
Date: Sun, 10 May 2015 21:48:08 +0200
From: Luis Rocha <luiscrocha@...il.com>
To: john-users@...ts.openwall.com
Subject: Re: Multiple Cores and GPU
Got it, thank you magnum!
On Sun, May 10, 2015 at 9:11 PM, magnum <john.magnum@...hmail.com> wrote:
> On 2015-05-10 19:57, Luis Rocha wrote:
>
>> I am trying to get john to use all CPU cores and 1 GPU together.
>>
>> Not sure how to do this but my current system has 4 cores and 1 GPU but
>> when I try to use the --fork suffix I would like to set that for CPU
>> device
>> use --fork=4 but for GPU --fork=1
>> Because when I try to use --fork=4, John starts a second instance running
>> on the GPU which is useless ....
>>
>> Is this possible?
>>
>
> No... well, almost: In the (few) cases where the OpenCL format using the
> CPU device is as fast as a plain CPU format, you could just use
> "--format=md5crypt-opencl --fork=2 -dev=all" and it would use one process
> for GPU and another process using all CPU cores via OpenCL.
>
> Other than that, I'd recomment either using the --node option to split the
> job in parts and run them separately, or simply run two different attacks
> (eg. one using mask mode on GPU and another using wordlist mode on CPUs).
>
> 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.
