_mm_alignr_epi8
prk ^= _mm_alignr_epi8(_mm_setzero_si128(), *rcon_rot, 15);
*rcon_rot = _mm_alignr_epi8(*rcon_rot, *rcon_rot, 15);
rk = _mm_alignr_epi8(rk, rk, 1);
rk = _mm_alignr_epi8(prk, prkhi, 8);
rk = _mm_alignr_epi8(prk, prkhi, 8);
mc = _mm_alignr_epi8(mc, mc, 12);