sm4: simd support for armv8, x86_64 #390
Conversation
zonyitoo
commented
Nov 3, 2023
zonyitoo
commented
- x86_64: aesni, avx2
- armv8: crypto-extension, neon
- x86_64: aesni, avx2 - armv8: crypto-extension, neon
|
Thank you! It's a relatively sizable PR, so it will take some time for us to properly review it. |
|
Hello @newpavlov , what do you think about this PR? |
| //! SM4 with X86 AES-NI instruction set | ||
| //! | ||
| //! Implementation was borrowed from <https://www.cnblogs.com/kentle/p/15826075.html> by kentle. |
There was a problem hiding this comment.
I found an English description of the approach being used here which helped me understand how this actually works:
https://github.com/mjosaarinen/sm4ni
The trick is to use affine transforms to emulate the SM4 S-Box with the AES S-Box. The S-Boxes are both based on finite field inversion, but use different affine transforms and even polynomial basis for the finite field. However, different polynomial bases are affine isomorphic.
Interesting approach!
|
@zonyitoo it looks pretty impressive but as @newpavlov said it is large and because of that hard to review, especially because it contains so many backends in a single PR. We're about to start making some breaking changes and I was looking through to see if we should land this PR first. I think it would probably make sense to try to land this as part of our next breaking release cycle. See #394 |
