Openwall GNU/*/Linux - a small security-enhanced Linux distro for servers
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Fri, 22 Jun 2012 02:12:16 +0200
From: Frank Dittrich <frank_dittrich@...mail.com>
To: john-dev@...ts.openwall.com
Subject: Re: relbench and changed format names

On 06/22/2012 01:37 AM, magnum wrote:
> Also, it seems icc precompiled intrinsics just make things slower here
> for nt2. Please run "make clean testpara32" and post the output.

Compiling assembler (4x) benchmarks
Benchmarking: Raw MD4 [128/128 SSE2 4x]... 11834K c/s real, 11954K c/s
virtual
Benchmarking: Raw MD5 [128/128 SSE2 4x]... 7835K c/s real, 7914K c/s virtual
Benchmarking: FreeBSD MD5 [32/32]... 4179 c/s real, 4179 c/s virtual
Benchmarking: Raw SHA-1 [128/128 SSE2 4x]... 3998K c/s real, 3998K c/s
virtual

Compiling md4 intrinsics benchmarks with PARA 1 (4x)
Benchmarking: Raw MD4 [128/128 SSE2 intrinsics 4x1]... 12808K c/s real,
12937K c/s virtual

Compiling md4 intrinsics benchmarks with PARA 2 (8x)
Benchmarking: Raw MD4 [128/128 SSE2 intrinsics 4x2]... 9862K c/s real,
9962K c/s virtual

Compiling md5 intrinsics benchmarks with PARA 1 (4x)
Benchmarking: Raw MD5 [128/128 SSE2 intrinsics 4x1]... 9129K c/s real,
9221K c/s virtual
Benchmarking: FreeBSD MD5 [128/128 SSE2 intrinsics 4x1]... 7972 c/s
real, 8052 c/s virtual

Compiling md5 intrinsics benchmarks with PARA 2 (8x)
Benchmarking: Raw MD5 [128/128 SSE2 intrinsics 4x2]... 8795K c/s real,
8884K c/s virtual
Benchmarking: FreeBSD MD5 [128/128 SSE2 intrinsics 4x2]... 7624 c/s
real, 7624 c/s virtual

Compiling sha1 [16x4] intrinsics benchmarks with PARA 1 (4x)
Benchmarking: Raw SHA-1 [128/128 SSE2 intrinsics 4x1]... 4152K c/s real,
4152K c/s virtual

Compiling sha1 [16x4] intrinsics benchmarks with PARA 2 (8x)
Benchmarking: Raw SHA-1 [128/128 SSE2 intrinsics 4x2]... 2904K c/s real,
2934K c/s virtual

Compiling sha1 [80x4] intrinsics benchmarks with PARA 1 (4x)
Benchmarking: Raw SHA-1 [128/128 SSE2 intrinsics 4x1]... 3576K c/s real,
3576K c/s virtual

Compiling sha1 [80x4] intrinsics benchmarks with PARA 2 (8x)
Benchmarking: Raw SHA-1 [128/128 SSE2 intrinsics 4x2]... 2836K c/s real,
2836K c/s virtual
======================================================
model name	: Intel(R) Atom(TM) CPU N270   @ 1.60GHz
gcc "gcc" version: gcc version 4.6.3 20120306 (Red Hat 4.6.3-2) (GCC)
Best -m32 paras:
MD4:	para_1 12937K c/s	asm 4x80 11954K c/s	
MD5:	para_1 9221K c/s	asm 4x80 7914K c/s	
MD5C:	para_1 8052 c/s	para_2 7624 c/s	asm 4x80 4179 c/s	
SHA1:	para_1 4x16 4152K c/s	asm 4x80 3998K c/s	para_1 4x80 3576K c/s	


Frank

Powered by blists - more mailing lists

Your e-mail address:

Powered by Openwall GNU/*/Linux - Powered by OpenVZ