[csw-maintainers] T4 crypto support in openssl 1.0
Jan Holzhueter
jh at opencsw.org
Thu May 24 11:11:53 CEST 2012
Hi,
Am 24.05.12 09:26, schrieb Yann Rouillard:
> Here is a test you could do:
>
> - launch the following commands with the current openssl:
> openssl speed -evp aes-128-cbc
> openssl speed -evp aes-192-cbc
> openssl speed -evp aes-256-cbc
>
> - then install openssl from my experimental repository and launch the
> following commands:
> openssl speed -engine t4 -evp aes-128-cbc
> openssl speed -engine t4 -evp aes-192-cbc
> openssl speed -engine t4 -evp aes-256-cbc
>
> then compare !
Ok here is my data:
I only used one test. It's impressive enough :)
Old Packages:
./openssl engine
(dynamic) Dynamic engine loading support
(pkcs11) PKCS #11 engine support
./openssl speed -evp aes-128-cbc
The 'numbers' are in 1000s of bytes per second processed.
type 16 bytes 64 bytes 256 bytes 1024 bytes 8192
bytes
aes-128-cbc 85871.89k 92796.71k 98383.53k 100017.49k
100256.43k
just for test if that helps:
./openssl speed -engine pkcs11 -evp aes-128-cbc
engine "pkcs11" set.
The 'numbers' are in 1000s of bytes per second processed.
type 16 bytes 64 bytes 256 bytes 1024 bytes 8192
bytes
aes-128-cbc 86177.13k 94786.86k 99174.31k 101733.38k
100578.65k
New packages:
./openssl engine
(t4) SPARC T4 engine support
(dynamic) Dynamic engine loading support
(pkcs11) PKCS #11 engine support
./openssl speed -evp aes-128-cbc
The 'numbers' are in 1000s of bytes per second processed.
type 16 bytes 64 bytes 256 bytes 1024 bytes 8192
bytes
aes-128-cbc 413248.17k 784163.48k 1008435.11k 1061075.63k
1103601.66k
./openssl speed -engine t4 -evp aes-128-cbc
engine "t4" set.
The 'numbers' are in 1000s of bytes per second processed.
type 16 bytes 64 bytes 256 bytes 1024 bytes 8192
bytes
aes-128-cbc 410416.72k 781939.48k 999663.45k 1055936.51k
1103432.36k
So same here T4 engine is enabled by default. Well according to this post:
https://blogs.oracle.com/darren/entry/howto_turn_off_sparc_t4
You can't turn it off anyway.
it seems to work :)
Greetings
Jan
More information about the maintainers
mailing list