Thanks in advance for your help.
Robert L. Scott
Application Engineer
Platform and Technology Enabling - Transactional e-Business
Intel Corporation
O: 505-893-1292
C: 505-681-7097
______________________________________________________________________
OpenSSL Project http://www.openssl.org
Development Mailing List opens...@openssl.org
Automated List Manager majo...@openssl.org
Errr ... is there a version of Win64 available for OpenSSL developers?
Cheers,
Ben.
--
http://www.apache-ssl.org/ben.html
"There is no limit to what a man can do or how far he can go if he
doesn't mind who gets the credit." - Robert Woodruff
Regarding Itanium. I've just got some performance numbers for 'openssl
speed rsa dsa' on a 733MHz Itanium box running Linux. Here is vanilla
0.9.6a compiled with gcc version 2.96 20000731 (Red Hat Linux 7.1
2.96-81):
sign verify sign/s verify/s
rsa 512 bits 0.0036s 0.0003s 275.3 2999.2
rsa 1024 bits 0.0203s 0.0011s 49.3 894.1
rsa 2048 bits 0.1331s 0.0040s 7.5 250.9
rsa 4096 bits 0.9270s 0.0147s 1.1 68.1
sign verify sign/s verify/s
dsa 512 bits 0.0035s 0.0043s 288.3 234.8
dsa 1024 bits 0.0111s 0.0135s 90.0 74.2
Here is 0.9.6a compiled with SGIcc version 0.01.0-12 (see
http://oss.sgi.com/projects/Pro64/)
sign verify sign/s verify/s
rsa 512 bits 0.0023s 0.0002s 431.4 5527.8
rsa 1024 bits 0.0103s 0.0005s 96.7 1874.9
rsa 2048 bits 0.0620s 0.0018s 16.1 566.1
rsa 4096 bits 0.4120s 0.0064s 2.4 157.3
sign verify sign/s verify/s
dsa 512 bits 0.0019s 0.0023s 518.1 435.8
dsa 1024 bits 0.0054s 0.0066s 185.8 151.9
And here is 0.9.6a with my assembler module (something to be proud over,
huh?):
sign verify sign/s verify/s
rsa 512 bits 0.0021s 0.0001s 486.7 9627.9
rsa 1024 bits 0.0055s 0.0002s 182.4 4481.1
rsa 2048 bits 0.0244s 0.0006s 40.9 1726.3
rsa 4096 bits 0.1295s 0.0018s 7.7 561.5
sign verify sign/s verify/s
dsa 512 bits 0.0012s 0.0013s 860.5 756.6
dsa 1024 bits 0.0023s 0.0028s 427.3 362.8
Note that performance improvement for 512-bit keys is less impressive
than for larger keys. That's because apparently there is an extra player
in the field, namely BN_from_montgomery(). When 'openssl speed rsa512'
(the one compiled with gcc!!!) is profiled, BN_from_montgomery appears
in second position with 16% result. The first position is taken by
bn_mul_add_words consuming 35% of CPU. 'openssl speed rsa1024' gives
different picture, bn_mul_add_words consumes 48%, while
BN_from_mongomery is moved to third position with 10% result.
Cheers, Andy.
Do you have some numbers from P3s or P2s in the 800mhz range? I'd love
to compare.
Also I'm curious, is the box running DDR or SDRAM? I keep hearing
wonderful things about DDR performance but I haven't seen any numbers.
- - M
- -----Original Message-----
From: Andy Polyakov [SMTP:ap...@fy.chalmers.se]
Sent: Friday, May 25, 2001 12:48 PM
To: opens...@openssl.org
Subject: Re: openssl for win64
-----BEGIN PGP SIGNATURE-----
Version: PGPfreeware 6.5.8 for non-commercial use <http://www.pgp.com>
iQA/AwUBOw6SQYwiya8i+ZOTEQLHGgCgjFAbdhAlvQYeMvtn9V+0tCp8PEkAoJyn
B+wkckNRNeR0UOAVYcFD4S4D
=OZpK
-----END PGP SIGNATURE-----
Here is 733MHz PIII:
sign verify sign/s verify/s
rsa 512 bits 0.0020s 0.0002s 501.0 5870.5
rsa 1024 bits 0.0099s 0.0005s 100.7 2016.9
rsa 2048 bits 0.0584s 0.0017s 17.1 605.4
rsa 4096 bits 0.3841s 0.0057s 2.6 175.8
sign verify sign/s verify/s
dsa 512 bits 0.0017s 0.0021s 577.9 484.8
dsa 1024 bits 0.0049s 0.0059s 204.6 169.0
> Also I'm curious, is the box running DDR or SDRAM?
Itanium box? Have no idea, I've never been even near it...
> ... on a 733MHz Itanium box running Linux.
> ...
> And here is 0.9.6a with my assembler module:
Note that the module is not committed yet. I have one last function to
implement, namely bn_div_words, and some comments to write.
> sign verify sign/s verify/s
> rsa 512 bits 0.0021s 0.0001s 486.7 9627.9
> rsa 1024 bits 0.0055s 0.0002s 182.4 4481.1
> rsa 2048 bits 0.0244s 0.0006s 40.9 1726.3
> rsa 4096 bits 0.1295s 0.0018s 7.7 561.5
> sign verify sign/s verify/s
> dsa 512 bits 0.0012s 0.0013s 860.5 756.6
> dsa 1024 bits 0.0023s 0.0028s 427.3 362.8
Andy.