Subject: Re: Damn slow RSA_generate_key
To: Martin Husemann <martin@duskware.de>
From: Gerald Richter <darklord@neonshadow.net>
List: port-sparc
Date: 01/30/2003 07:33:09
I suspect it has more to do with the arcitecture of the chips in 
question... the intel chips include faster hardware instructions useful 
for cryptography that just aren't present in sparcs and take forever to 
emulate in software... not to ignite a war here, just pointing that out. 
GCC, imo, does a great job when it optimizes for the hardware you are 
running on (686 code vs 386 code made a massive diff on my quad ppro 
server)...

Just my half a cent worth.
-Gerald

Martin Husemann wrote:

>On Thu, Jan 30, 2003 at 02:34:45PM +0100, Bernd Walter wrote:
>
>  
>
>>[103]cicely21# ./blah 
>>69.732u 0.129s 1:11.98 97.0%    0+0k 1+1io 0pf+0w
>>    
>>
>
>My 100 MHz Krups needs 26 seconds.
>My 300 MHz ultra sparc needs 6.59 s.
>My 400 MHz PII 0.89 s.
>
>So gcc sucks (for sparcs at least), sparcs are dog slow, old Intels get 7.5 x
>the performance at the same clock frequency.
>
>Which of these is a suprise for you?
>
>Martin
>
>  
>