search

intel / isa-l_crypto

Other
267 stars 80 forks source link

SM3 mb optimization for aarch64 by SVE #95

Closed daniel-hu-arm closed 2 years ago

daniel-hu-arm commented 2 years ago

This patch optimize the sm3-mb algorithm using the SVE feature of aarch64. SVE is optional feature for ARM v8

Tested on modern micro-architecture, it shows an 39% uplift in performance over neon

Signed-off-by: Daniel Hu Daniel.Hu@arm.com

gbtucker commented 2 years ago

Is there more of these coming?

daniel-hu-arm commented 2 years ago

yes, will have more. but perhaps we will hold until the pending PRs are cleared.

gbtucker commented 2 years ago

integrated