Re:Base64 encoding and decoding at almost the speed of a mem...
by vardump on Wednesday November 06, 2019 @01:02PM
(#21465104)
Yeah, missed that you can now fit 64 8-bit entries in a single 512 bit register!
Generally my point stands that LUTs are not vectorizable, except for small special cases like this.
This is doing exactly that with `vpermb` and `vpermi2b`.