[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20130516112819.GA4300@openwall.com>
Date: Thu, 16 May 2013 15:28:19 +0400
From: Solar Designer <solar@...nwall.com>
To: john-dev@...ts.openwall.com
Subject: Re: password generation on GPU
Hi Sayantan,
On Thu, May 16, 2013 at 12:48:50PM +0530, Sayantan Datta wrote:
> In this implementation of mask-mode the key is generated outside the
> format. However in the PG-test branch we were generating the keys inside
> the format using GPU . So how do I start implementing the mask mode? Should
> we generate the keys globally outside the format using a GPU ?
Oh, at first I misread what you wrote above. You're asking not whether
to generate keys on host vs. GPU (the short answer to which is that we
should do both of these things), but whether to do the GPU portion of
key generation inside the format or with code shared between formats.
So far, we only considered doing it from the same kernels that do the
crypto, so we'd have duplicate implementations of the GPU side of mask
mode (one for each fast hash format). I think that for mask mode this
is in fact what we should be doing, and those duplicate implementations
will end up having enough format-specific optimizations for the code
"duplication" to be worth it. (Ditto for code responsible for comparing
computed vs. loaded hashes, using bitmaps and such.) We may also have
generic implementation(s) (or portions thereof), to be #include'd from
multiple per-format OpenCL kernels. Using separate kernels (with
communication via global memory?) is likely slower.
Thanks,
Alexander
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.
