Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Wed, 29 Aug 2012 11:55:14 -0400 (EDT)
From: "Brad Tilley" <brad@...ystems.com>
To: john-users@...ts.openwall.com
Subject: Re: Questions about compiling for Optimal CPU 
     Performance

Hi Magnum,

>> My older Intel Celeron CPU lists sse, sse2 and sse3 as supported in
>> /proc/cpuinfo, should I make-x86-sse2 or make-x86-native when compiling
>> jumbo-6? Or does the difference really matter?
>
> You should probably use linux-x86-64i. If you run a 32-bit OS, you should
> really ask yourself why.

You're joking, aren't you? Some CPUs are still 32-bit only or have 64-bit
capability, but disable it (such as some Intel Atoms). So when asked why,
that is the reason I sometimes use a 32-bit operating system and cannot
use a 64-bit OS for those specific CPUs. You can understand that I hope.
Not sure why this would be questioned... am I missing something? An inside
joke only related to JtR perhaps?

>> Also, the Makefile has two
>> possibilities for OpenMP...
>>
>> 1. -fopenmp
>> 2. -fopenmp -msse2
>>
>> Which one should be uncommented for optimal performance when the CPU
>> lists
>> multiple sse implementations. For example, should a CPU support sse4,
>> would I be limiting performance if I opted for OpenMP with -msse2, or
>> not?
>
> Some 32-bit builds need the latter. For a 64-bit build they produce the
> same result.

I find the former to work better than the latter on 32-bit systems.

Brad


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.