ppc_AES_[en|de]crypt perform at 18 cycles per byte processed with 128-bit key, which is ~40 better than 64-bit code generated by gcc 4.0.

-ppc_AES_[en|de]crypt perform at 18 cycles per byte processed with 128-bit key, which is ~40　better than 64-bit code generated by gcc 4.0.