On Wednesday, December 12, 2018 at 4:38:56 AM UTC-5, Jeffrey Walton wrote:
>
> Hi Everyone,
>
> We cut-in x25519 key exchange this week. It is based on Adam Langley's 
> work at https://github.com/agl/curve25519-donna .
> ...
>
> Also see https://github.com/weidai11/cryptopp/issues/761 .
>

The integration is complete and Issue 761 was closed this morning.

We switched to Andrew Moon's implementation at the last minute. Moon's 
implementation was slightly faster and public domain; but it was a tad bit 
more difficult to integrate.

We provided a SSE2 implementation but disabled it because it was not 
profitable. You can enable it if needed by defining CRYPTOPP_CURVE25519_SSE2 
in donna.h.

The numbers for the implementations when tested on a modern Core i5-6400 
@2.7 GHz in milliseconds/operation:

   * Langley, C++, 0.050
   * Moon, C++: 0.040
   * Moon, SSE2: 0.061
   * Moon, native: 0.045

"Moon, native" was a test using "-march=native" to see if AVX made a 
difference. Langley performs about 20K agreements per second, while Moon 
performs about 25K agreements per second.

Jeff

-- 
You received this message because you are subscribed to "Crypto++ Users". More 
information about Crypto++ and this group is available at 
http://www.cryptopp.com and 
http://groups.google.com/forum/#!forum/cryptopp-users.
--- 
You received this message because you are subscribed to the Google Groups 
"Crypto++ Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to