Actions
Feature #1808
openhardware crypto acceleration feature using aes
Status:
New
Priority:
Normal
Assignee:
-
Category:
-
Start date:
2011-11-23
Due date:
% Done:
0%
Estimated time:
Difficulty:
Medium
Tags:
needs-triage
Gerrit CR:
Description
A placeholder request...
The latest openssl from you know who added aes_ni to support the on chip crypto acceleration.
see http://blogs.oracle.com/darren/entry/openssl_versions_in_solaris
for information, seeing this followup blog:
http://blogs.oracle.com/darren/entry/howto_turn_off_sparc_t4
I tried first on my 32-bit AMD 2500 Barton running OI_151a:
$ isainfo -v 32-bit i386 applications ahf sse fxsr amd_3dnowx amd_3dnow amd_mmx mmx cmov sep cx8 tsc fpu $ isalist pentium_pro+mmx pentium_pro pentium+mmx pentium i486 i386 i86 type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes aes-128-cbc 36983.19k 65556.87k 82042.69k 87296.53k 89114.18k
and on a 64-bit AMD Opteron single socket 6128 :
$ isainfo -v 64-bit amd64 applications amd_svm amd_lzcnt popcnt amd_sse4a tscp ahf cx16 sse3 sse2 sse fxsr amd_3dnowx amd_3dnow amd_mmx mmx cmov amd_sysc cx8 tsc fpu 32-bit i386 applications amd_svm amd_lzcnt popcnt amd_sse4a tscp ahf cx16 sse3 sse2 sse fxsr amd_3dnowx amd_3dnow amd_mmx mmx cmov amd_sysc cx8 tsc fpu $ isalist amd64 pentium_pro+mmx pentium_pro pentium+mmx pentium i486 i386 i86 ... type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes aes-128-cbc 55081.64k 89261.27k 110092.46k 116614.49k 118759.42k
comparatively speaking, (with the blog), performance without hardware acceleration, even on a {almost} ten year old 32-bit processor, isn't too shabby compared with SPARC.
Can't wait for our imminent AMD Bulldozer upgrade which will have 'aes' support. Hence this request for a solution on the illumos platforme.
No data to display
Actions