Hi Herbert and others,
attached is a patch with support for VIA C7 crypto engine providing
SHA1/SHA256 digests. It compiles into a new module padlock-sha.ko.
Currently it allocates 1 page for its buffer and if there are more data
to be hashed it falls back to software SHA implementation. By defau
On Wed, Jul 05, 2006 at 04:57:55PM +1200, Michal Ludvig wrote:
>
> Here are some benchmarks from tcrypt.ko (mode=303 for sha1 and mode=304
> for sha256):
Looks impressive!
The only thing I'd like to tweak is to use the newly added template
mechanism to pick the fallback. But you don't have to w
On Wed, Jul 05, 2006 at 03:21:41PM +1000, herbert wrote:
>
> Also, we can probably bring down the 16-byte numbers if we change the
> digest layer to cater for a direct digest interface that takes one sg
> entry, i.e., if it's one sg entry then feed it directly to the algo's
> digest function, othe