[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <4EBD8A61.8010605@hushmail.com>
Date: Fri, 11 Nov 2011 21:49:37 +0100
From: magnum <john.magnum@...hmail.com>
To: john-dev@...ts.openwall.com
Subject: Re: more targets using sse-intrinsics.S
These problems must have been there with the sse2i targets all the time.
Would it be possible to build with both mmx-md5.S and
sse-intrinsics-32.S as an interim fix? Maybe that's not less work...
magnum
2011-11-11 21:45, jfoug wrote:
> Here are some things we will need to work on. These timings are against SSE2 built with mmx-md5.S and sse-intrisic-32.S
>
> SSE2 mmx-md5.S
> Benchmarking: HMAC MD5 [hmac-md5 SSE2]... DONE
> Raw: 3743K c/s
> Benchmarking: MS-SQL [mssql SSE2]... DONE
> Many salts: 11173K c/s Only one salt: 6388K c/s
> Benchmarking: MS-SQL05 [ms-sql05 SSE2]... DONE
> Many salts: 11347K c/s
> Only one salt: 7660K c/s
> Benchmarking: MySQL 4.1 double-SHA-1 [mysql-sha1 SSE2]... DONE
> Raw: 4650K c/s
> Benchmarking: PIX MD5 [pix-md5 SSE2]... DONE
> Raw: 10070K c/s
>
> SSE2 sse-intrisic-32.S (simply reverts back to 'any')
> Benchmarking: HMAC MD5 [hmac-md5]... DONE
> Raw: 1138K c/s
> Benchmarking: MS-SQL [mssql]... DONE
> Many salts: 3354K c/s Only one salt: 2964K c/s
> Benchmarking: MS-SQL05 [ms-sql05]... DONE
> Many salts: 3284K c/s
> Only one salt: 3100K c/s
> Benchmarking: MySQL 4.1 double-SHA-1 [mysql-sha1]... DONE
> Raw: 1705K c/s
> Benchmarking: PIX MD5 [pix-md5]... DONE (NOTE, we have a thin format, we should use for this one).
> Raw: 3614K c/s
>
> So, these formats:
>
> HMAC MD5
> MS-SQL
> MS-SQL05
> MySQL
> Pix-md5
>
> Have sse2 code, but ONLY the 32bit mmx-md5.S code. We need to port these to intrinsic, and they will get huge performance gains over many systems (all 64 bit, I assume).
>
> These 2 also need looked at:
>
> SSE2 mmx-md5.S
> Benchmarking: Netscape LDAP SSHA [4x]... DONE
> Many salts: 10398K c/s
> Only one salt: 6919K c/s
> Benchmarking: Netscape LDAP SHA [4x]... DONE
> Raw: 7860K c/s
>
> SSE2 sse-intrisic-32.S (simply reverts back to 'any')
> Benchmarking: Netscape LDAP SSHA [8x]... DONE
> Many salts: 3133K c/s
> Only one salt: 2954K c/s
> Benchmarking: Netscape LDAP SHA [8x]... DONE
> Raw: 3173K c/s
>
> I am not sure about these. I do not know what the [8x] is, or where it comes from.
>
>
> Jim.
>
>
>
>
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.
