[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <e9ffca014d09121b02311f4951727920@smtp.hushmail.com>
Date: Thu, 19 Feb 2015 23:08:47 +0100
From: magnum <john.magnum@...hmail.com>
To: john-dev@...ts.openwall.com
Subject: Re: descrypt speed
On 2015-02-19 22:06, Sayantan Datta wrote:
> On Fri, Feb 20, 2015 at 12:34 AM, magnum <john.magnum@...hmail.com> wrote:
>
>> I did some testing on a GTX980 using (1,1). Tried it with the good old
>> Gawker dataset, with 3844 salts. After all kernels were compiled (yeah
>> that took a while) and running full speed, virtual memory footprint was
>> 38.5 GB(!) but resident size was just 2.19 GB which is pretty sane.
>>
>
> OMG 38.5 GB!! According to Royce, binary size is around 300KB, so the
> expected memory footprint should be around 300KB x 3844 = 1.1GB. Maybe this
> means unlike AMD, the binaries does not contain the ISA, which probably is
I do not think the VIRT figure tells us much: I tried loading (using
0,0) just 5 hashes to a 7970 on Super, and VIRT was 144 GB (!) while RES
was just 140 MB. Using the Titan, same VIRT and 172 MB RES. Even the CPU
device show similar figures. And other formats does as well. The
resident size is what matters.
> built from the binaries and is much larger than 300KB. Did you try running
> it the second time? Was the compilation any quicker, assuming the binaries
> weren't deleted ? Also what speeds are you getting with multiple-salts ?
Well it was quicker 2nd time but the nvidia driver already caches for us
(to ~/.nv/ComputeCache) so I'm not sure the caching you added makes much
of a difference.
>> I also did some changes to ensure proper Makefile dependencies for
>> opencl_DES_hst_dev_shared.h, so you can now edit that file and be sure a
>> simple "make" really rebuilds all files involved.
>>
>
> Thank You again. We could do the switching at runtime, but prior to that we
> need to test devices which can really benefit form (1,1) kernel.
Due to the awfully slow loading we might want it as a john.conf option
instead, or an environment variable (or both).
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.
