Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <f8000ee73ae39a0bcd6d6c1e631b860c@smtp.hushmail.com>
Date: Wed, 7 Oct 2015 10:14:48 +0200
From: magnum <john.magnum@...hmail.com>
To: john-users@...ts.openwall.com
Subject: Re: nVidia Maxwell support (especially descrypt)?

On 2015-10-07 09:16, Sayantan Datta wrote:
> On Wed, Oct 7, 2015 at 8:44 AM, Solar Designer <solar@...nwall.com> wrote:
>> Previously, for best performance, we'd use HARDCODE_SALT = 1 and
>> FULL_UNROLL = 1 for tahiti. Setting only HARDCODE_SALT, results in much
>> smaller sized binaries but also looses some performance on tahiti.
>
> PARALLEL_BUILD which I thought would be faster by large margin, isn't so.
> Seems like opencl compilers are still sequential. Otherwise on our side,
> this bottleneck has been eliminated.
>
>> but enabling it made the compiler crash with a
>> weird error (when targeting Titan X), so maybe something isn't MT-safe.
>
> This is new. I'll look into.

FWIW I did not see any problem when testing it briefly (I intend to test 
more). However, I also saw no (or very little) speedup. I was hoping for 
similar significant speedup as "make -sj32" when buildin john itself :-(

On a side note, zero binaries does not work so I used this:

perl -e '$c64 = 
"./0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz"; 
foreach $c1 (split //, $c64) { foreach $c2 (split //, $c64) { print 
"$c1${c2}Nf8Sbh3HDfQ\n"; } }' > pw-fakedes

>> Really need documentation for this, even if it's still in development.
>> And then we could proceed to discuss how to revise it to make it more
>> usable.
>>
>
> Sure. We'll do.

>> And of course we'll also need to include some LOP3.LUT S-boxes.
>> If Roman's are still unreleased (except for S4), then Janet's.

FWIW the needed inline stuff is already present (and tested) and I just 
refactored it from lop3_lut() to lut3() like what's being used in 
Roman's s4.

After including opencl_misc.h you can now use:

#if HAVE_LUT3
[code using "r = lut3(a, b, c, t);"]
#else
(...)

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.