Re: VIA C3 AES code ?

From: Mike Tancsa <mike_at_sentex.net>
Date: Fri, 27 Aug 2004 14:28:44 -0400
At 04:32 PM 10/08/2004, Mark Murray wrote:
> > It looks like OpenSSL has this recently committed into their tree
> >
> > http://marc.theaimsgroup.com/?l=openssl-cvs&m=109148330200378&w=2
> > which came from
> > http://www.logix.cz/michal/devel/padlock/
>
>Yeah. I'm watching OpenSSL with interest.

Hi,
         Thanks for bringing this code into FreeBSD!  Just testing it out 
on one of the new EPIA boards. It certainly seems to be using the CPU's 
extra instructions, but when you ask openssl, RNG is not hooked up.  Just 
curious as to why ?

Speed looks great!
type             16 bytes     64 bytes    256 bytes   1024 bytes   8192 bytes
aes-256-ecb      66503.60k   248582.69k   649266.49k  1071507.80k  1329273.39k
aes-256-ecb       8320.88k     8661.14k     8804.91k     8812.70k     8821.30k

Now to try and hack gbde support :)

53beta# openssl engine padlock
(padlock) VIA PadLock (no-RNG, ACE)
53beta#

Yet,
53beta# dmesg | grep ^CPU
CPU: VIA C3 Nehemiah+RNG+ACE (1199.79-MHz 686-class CPU)
53beta#

53beta# openssl speed -evp aes-256-ecb
To get the most accurate results, try to run this
program when this computer is idle.
Doing aes-256-ecb for 3s on 16 size blocks: 1559132 aes-256-ecb's in 3.00s
Doing aes-256-ecb for 3s on 64 size blocks: 406739 aes-256-ecb's in 3.01s
Doing aes-256-ecb for 3s on 256 size blocks: 103109 aes-256-ecb's in 3.00s
Doing aes-256-ecb for 3s on 1024 size blocks: 25868 aes-256-ecb's in 3.01s
Doing aes-256-ecb for 3s on 8192 size blocks: 3218 aes-256-ecb's in 2.99s
OpenSSL 0.9.7d 17 Mar 2004
built on: Fri Aug 27 04:49:37 EDT 2004
options:bn(64,32) md2(int) rc4(idx,int) des(ptr,risc1,16,long) aes(partial) 
blowfish(idx)
compiler: cc
available timing options: USE_TOD HZ=128 [sysconf value]
timing function used: getrusage
The 'numbers' are in 1000s of bytes per second processed.
type             16 bytes     64 bytes    256 bytes   1024 bytes   8192 bytes
aes-256-ecb       8320.88k     8661.14k     8804.91k     8812.70k     8821.30k
53beta#
vs
53beta# openssl speed -evp aes-256-ecb -engine padlock
engine "padlock" set.
To get the most accurate results, try to run this
program when this computer is idle.
Doing aes-256-ecb for 3s on 16 size blocks: 12453020 aes-256-ecb's in 3.00s
Doing aes-256-ecb for 3s on 64 size blocks: 11644992 aes-256-ecb's in 3.00s
Doing aes-256-ecb for 3s on 256 size blocks: 7622561 aes-256-ecb's in 3.01s
Doing aes-256-ecb for 3s on 1024 size blocks: 3145295 aes-256-ecb's in 3.01s
Doing aes-256-ecb for 3s on 8192 size blocks: 482357 aes-256-ecb's in 2.97s
OpenSSL 0.9.7d 17 Mar 2004
built on: Fri Aug 27 04:49:37 EDT 2004
options:bn(64,32) md2(int) rc4(idx,int) des(ptr,risc1,16,long) aes(partial) 
blowfish(idx)
compiler: cc
available timing options: USE_TOD HZ=128 [sysconf value]
timing function used: getrusage
The 'numbers' are in 1000s of bytes per second processed.
type             16 bytes     64 bytes    256 bytes   1024 bytes   8192 bytes
aes-256-ecb      66503.60k   248582.69k   649266.49k  1071507.80k  1329273.39k
53beta# 
Received on Fri Aug 27 2004 - 16:24:17 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:08 UTC