Andy Polyakov wrote:
| 
| [...]
| It might happen that RC4_INT=char would perform better on plain P4 as
| well... It's never simple, huh?

The fact that RC4_INT=long is slower on Nocona may be explained by the
small 16 kB L1 data cache of this processor (to be compared with 64 kB
on AMD processors). When RC4_INT=long, the RC4 table occupies 256*8 =
2 kB in memory, this probably forces some data out of L1 cache... But,
heh, maybe it's not that simple !

| [...]
| All rc4-amd64s rely on partial register arithmetics, something P4 core 
| was never good at [which is why you see "How IA32E would perform remains 
| to be seen..." in commentary section]...

I thought P4 were pretty good on partial register arithmetic. Hmm, my
source was unreliable :)

-- 
Marc Bevand                              http://epita.fr/~bevand_m
Computer Science School EPITA - System, Network and Security Dept.
______________________________________________________________________
OpenSSL Project                                 http://www.openssl.org
Development Mailing List                       [EMAIL PROTECTED]
Automated List Manager                           [EMAIL PROTECTED]

Reply via email to