Accelerate GCM with PCLMULQDQ-NI

Accelerate GCM with PCLMULQDQ-NI

[Huang Ying]

[-- Attachment #1: Type: text/plain, Size: 621 bytes --]

Hi,

In addition to AES-NI, Intel will provide PCLMULQDQ-NI (carry-less
multiplication) in future CPU to accelerate GCM mode. The document can
be found here:

http://software.intel.com/en-us/articles/carry-less-multiplication-and-its-usage-for-computing-the-gcm-mode/

It follows the same design as that of AES-NI, that is, XMM registers are
used.

To accelerate GCM with it, I make the following design:

1. Implement ghash as an ahash algorithm, Use ghash in gcm
implementation.
2. Provide a new implementation of ghash with PCLMULQDQ-NI.

What do you think about that?

Best Regards,
Huang Ying


[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 197 bytes --]

Re: Accelerate GCM with PCLMULQDQ-NI

[Herbert Xu]

On Wed, Mar 18, 2009 at 04:52:12PM +0800, Huang Ying wrote:
>
> To accelerate GCM with it, I make the following design:
> 
> 1. Implement ghash as an ahash algorithm, Use ghash in gcm
> implementation.
> 2. Provide a new implementation of ghash with PCLMULQDQ-NI.
> 
> What do you think about that?

Sounds OK to me.  I presume the choice of ahash instead of shash
is due to the use of XMM registers, just like AES?

Cheers,
-- 
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <herbert@gondor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt

Re: Accelerate GCM with PCLMULQDQ-NI

[Huang Ying]

[-- Attachment #1: Type: text/plain, Size: 669 bytes --]

On Sun, 2009-03-29 at 15:43 +0800, Herbert Xu wrote:
> On Wed, Mar 18, 2009 at 04:52:12PM +0800, Huang Ying wrote:
> >
> > To accelerate GCM with it, I make the following design:
> > 
> > 1. Implement ghash as an ahash algorithm, Use ghash in gcm
> > implementation.
> > 2. Provide a new implementation of ghash with PCLMULQDQ-NI.
> > 
> > What do you think about that?
> 
> Sounds OK to me.  I presume the choice of ahash instead of shash
> is due to the use of XMM registers, just like AES?

Yes, because PCLMULQDQ touches XMMs too. Just like AES, there is a raw
shash: ghash-clmul and an ahash via cryptd(ghash-clmul).

Best Regards,
Huang Ying


[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 197 bytes --]