I have the two necessary options in my kernel config. It also shows up on dmesg:
It shows up when I check [CMD="openssl"]engine[/CMD]
But when I run benchmarks to make sure it's working:
The speed doesn't change when the engine is enabled? Any idea what's going on here, and how to fix it? Ultimately I want to use it with OpenVPN. OpenVPN was not running during these tests, to ensure nothing else was using the padlock engine.
Additional information:
Edit: I also tested with SHA1 and SHA256, and again, there was no difference in speed with or without the -engine padlock.
Edit2: If no one has any ideas, I think this may be a bug. There has been, I believe, one commit that is padlock related to 8-STABLE, but I'm not sure if it has any bearing on what I'm experiencing. Unfortunately the two machines I have that have VIA processors and are padlock capable are in production, so I can't do much testing on this issue.
Code:
CPU: VIA Esther processor 1500MHz (1496.26-MHz 686-class CPU)
Origin = "CentaurHauls" Id = 0x6a9 Family = 6 Model = a Stepping = 9
Features=0xa7c9bbff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,CMOV,PAT,CLFLUSH,ACPI,MMX,FXSR,SSE,SSE2,TM,PBE>
Features2=0x181<SSE3,EST,TM2>
VIA Padlock Features=0x3fcc<RNG,AES,AES-CTR,SHA1,SHA256,RSA>
...
padlock0: <AES-CBC,SHA1,SHA256> on motherboard
Code:
(padlock) VIA PadLock (no-RNG, ACE)
(dynamic) Dynamic engine loading support
Code:
[I]With padlock:[/I]
openssl speed -engine padlock aes-128-cbc
[B]engine "padlock" set.[/B]
To get the most accurate results, try to run this
program when this computer is idle.
Doing aes-128 cbc for 3s on 16 size blocks: 2629077 aes-128 cbc's in 2.98s
Doing aes-128 cbc for 3s on 64 size blocks: 679466 aes-128 cbc's in 2.98s
Doing aes-128 cbc for 3s on 256 size blocks: 171106 aes-128 cbc's in 2.97s
Doing aes-128 cbc for 3s on 1024 size blocks: 43083 aes-128 cbc's in 2.98s
Doing aes-128 cbc for 3s on 8192 size blocks: 5386 aes-128 cbc's in 2.98s
OpenSSL 0.9.8q 2 Dec 2010
built on: date not available
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
[B]aes-128 cbc 14130.37k 14604.70k 14766.23k 14811.74k 14819.24k[/B]
[I]Without padlock:[/I]
openssl speed aes-128-cbc
To get the most accurate results, try to run this
program when this computer is idle.
Doing aes-128 cbc for 3s on 16 size blocks: 2629105 aes-128 cbc's in 2.98s
Doing aes-128 cbc for 3s on 64 size blocks: 679305 aes-128 cbc's in 2.98s
Doing aes-128 cbc for 3s on 256 size blocks: 171849 aes-128 cbc's in 2.98s
Doing aes-128 cbc for 3s on 1024 size blocks: 43087 aes-128 cbc's in 2.98s
Doing aes-128 cbc for 3s on 8192 size blocks: 5388 aes-128 cbc's in 2.98s
OpenSSL 0.9.8q 2 Dec 2010
built on: date not available
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
[B]aes-128 cbc 14107.15k 14604.34k 14770.01k 14810.18k 14820.70k[/B]Additional information:
Code:
[CMD="uname"]-mrs[/CMD]
FreeBSD 8.2-RELEASE-p2 i386
[CMD="which"]openssl[/CMD]
/usr/bin/openssl
[CMD="openssl"]version[/CMD]
OpenSSL 0.9.8q 2 Dec 2010Edit: I also tested with SHA1 and SHA256, and again, there was no difference in speed with or without the -engine padlock.
Edit2: If no one has any ideas, I think this may be a bug. There has been, I believe, one commit that is padlock related to 8-STABLE, but I'm not sure if it has any bearing on what I'm experiencing. Unfortunately the two machines I have that have VIA processors and are padlock capable are in production, so I can't do much testing on this issue.