dead hachers! Background: Maybe you know mmx packing/unpacking instruction set, what howewer works on bytes and larger volumes, like shorts, words, etc.

Quesion: Are there any instructions for any x86_64 processor, to do the similar packing/unpacking but only for bits?

Purpose: It is an idea for optimizing the DES/AES bit shifts, but if you have other proposals about it, you may answer as well.

Thanks a lot for answers