Hi Gilad,
On 12/22/21 2:59 PM, Gilad Ben-Yossef wrote:
On Wed, Dec 22, 2021 at 6:50 AM Tianjia Zhang
<tianjia.zhang@xxxxxxxxxxxxxxxxx> wrote:
Stand-alone implementation of the SM3 algorithm. It is designed
to have as little dependencies as possible. In other cases you
should generally use the hash APIs from include/crypto/hash.h.
Especially when hashing large amounts of data as those APIs may
be hw-accelerated. In the new SM3 stand-alone library,
sm3_compress() has also been optimized, instead of simply using
the code in sm3_generic.
I have a really minor nitpick: the commit message talks about changes
to sm3_compress() which was there in the original code but there is no
such function in the current code which is in a different patch and
file, so if you do another iteration for other reason, perhaps change
the commit message to refer to sm3_transform() instead? it's not
really important enough to warrant a new iteration on it's own...
Otherwise, I'm not smart enough to evaluate the changes to
sm3_transform() cryptographically but the overall approach of moving
to a standalone library seems sane to me.
So, for what it's worth -
Reviewed-by: Gilad Ben-Yossef <gilad@xxxxxxxxxxxx>
Gilad
Thanks for your suggestion. I agree with you. In the implementation of
sm3_generic, sm3_compress() is a sub-function of sm3_transform(). The
optimization is indeed for sm3_transform(). I will fix it in the next patch.
Best regards,
Tianjia
