Openwall GNU/*/Linux - a small security-enhanced Linux distro for servers
[<prev] [next>] [day] [month] [year] [list]
Date: Wed, 07 Mar 2012 21:38:13 +0100
From: magnum <john.magnum@...hmail.com>
To: john-dev@...ts.openwall.com
Subject: ALGORITHM_NAME et al in GPU formats

I believe you are using ALGORITHM_NAME incorrectly in most GPU formats.
The name is a little unintuitive: It's supposed to tell
implementation-specific things like 64/64 or SSE2 12x. In your case I
believe it should simply say CUDA or OpenCL unless there's something
more specific to tell.

The algorithm, salting etc can be freely described in FORMAT_NAME while
BENCHMARK_COMMENT can be used for things like (rounds=5000) - especially
for formats where this is not fixed.

Examples attached. One reason these are important is that once we
release a Jumbo with some FORMAT_NAME, changing it will confuse relbench
so it can't boast about future boost ;-)

magnum

View attachment "algo.diff" of type "text/x-patch" (1499 bytes)

Powered by blists - more mailing lists

Your e-mail address:

Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.