[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Fri, 15 Jun 2012 22:07:58 +0200
From: magnum <john.magnum@...hmail.com>
To: john-dev@...ts.openwall.com
Subject: Re: Re: Re: [patch] optional new raw sha1 implemetation
On 2012-06-15 21:13, magnum wrote:
> On 2012-06-15 20:02, Tavis Ormandy wrote:
>> 5% improvement is nice, but I'm dissapointed it's not higher. Could
>> you send
>> me a .o file? I would like to disassemble it and see where GCC goes
>> wrong,
>> hopefully there's a simple way to fix it.
>>
>> Version 3 attached.
>
> Tavis,
>
> I have committed this to magnum-jumbo for now, although it won't make it
> into a Jumbo release in the current form (magnum-jumbo is not really an
> official repo, but in the end most of it is likely to be the next Jumbo
> release).
>
> The current plugin "system" can't handle requirements like SSSE4, OpenCL
> >= x.y or CUDA/OpenCl. So any format that has such prerequistes are
> currently non-plugins (the only difference is two lines + #ifdefs in
> john.c, see eg. rawSHA256_fmt.c for an example).
I now turned the format into a non-plugin, depending on __SSE4_1__.
This renews my thoughts that we should have native build targets, eg.
linux-x86-64-native, that includes -march=native or whatever is needed
to enable everything available.
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.
