The cr.yp.to microblog: 2013.04.07 05:12:59

2013.04.07 05:12:59 (320735929818042368) from Daniel J. Bernstein, replying to "Diego F. Aranha 🕷️ (@dfaranha)" (320240157196435456):

@dfaranha @nickm_tor Yes, it's available as crypto_stream/aes128ctr/neon in SUPERCOP. <19 cycles/byte on Cortex A8.

Context

2013.04.02 03:37:48 (318900037524525056) from "Diego F. Aranha 🕷️ (@dfaranha)":

@nickm_tor I would check on NaCl first: http://nacl.cr.yp.to

2013.04.02 04:09:03 (318907904004591616) from "Nick Mathewson (@nickm_tor)", replying to "Diego F. Aranha 🕷️ (@dfaranha)" (318900037524525056):

@dfaranha nacl-2011 doesn't have a ARM AES implementation AFAICT. Is its C implementation speed-competitive?

2013.04.05 20:22:57 (320240157196435456) from "Diego F. Aranha 🕷️ (@dfaranha)", replying to "Nick Mathewson (@nickm_tor)" (318907904004591616):

@nickm_tor Sorry, I thought that their "NEON Crypto" paper covered AES as well. :(