vqtbl1q_u8
m = vqtbl1q_u8(vld1q_u8(in + size - AEGIS_BLOCK_SIZE),
vqtbl1q_u8(m ^ s, vld1q_u8(permute + size)));
vqtbl1q_u8(m, vld1q_u8(permute + size)));
v = vqtbl1q_u8(~vceqq_u8(v, vld1q_u8(tag_xor)),
w = vqtbl1q_u8(w, vld1q_u8(shift_rows));
w ^= vqtbl1q_u8(v ^ w, vld1q_u8(ror32by8));
vx = vqtbl1q_u8(qm0, vandq_u8(vx, x0f));
vy = vqtbl1q_u8(qm1, vy);
vx = vqtbl1q_u8(pm0, vandq_u8(px, x0f));
vy = vqtbl1q_u8(pm1, vy);
vx = vqtbl1q_u8(qm0, vandq_u8(vx, x0f));
vy = vqtbl1q_u8(qm1, vy);