Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20150917153601.GA16659@openwall.com>
Date: Thu, 17 Sep 2015 18:36:01 +0300
From: Solar Designer <solar@...nwall.com>
To: john-dev@...ts.openwall.com
Subject: wordlist mgetl() (was: Judy array)

On Thu, Sep 17, 2015 at 04:15:40PM +0200, magnum wrote:
> Here's a theory: Does the E5420 have a significant penalty for unaligned 
> SIMD loads?

Yes.

> If so, you should try this (wordlist.c:140)
> 
> -#if defined(SIMD_COEF_32) && !defined(_MSC_VER) && \
> +#if 0 && defined(SIMD_COEF_32) && !defined(_MSC_VER) && \
> 
> The alternative code might be faster.

I've just tried, with and without mmap.  I am getting similar speeds as
before (a sub-second difference one way or the other).  mmap is still
slower than pre-loading (with -mem=0).  That's for the 29M testcase;
maybe I should test on something more focused on wordlist performance.

BTW, that #if's handling of __MIC__ should probably be simplified - we
may write "!__MIC__" alongside "!__ALTIVEC__ && !__ARM_NEON__" rather
than inside the AVX-512 sub-expression.

Alexander

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.