vaddq_u32
ctr = vaddq_u32(ctr, ctr32_inc);
ctr = vaddq_u32(ctr, ctr32_inc);
ctr = vaddq_u32(ctr, ctr32_inc);
r0 = vaddq_u32(r0, r1); r3 ^= r0; r3 = rol16(r3);
r2 = vaddq_u32(r2, r3); r1 ^= r2; r1 = rol12(r1);
r0 = vaddq_u32(r0, r1); r3 ^= r0; r3 = rol8(r3);
r2 = vaddq_u32(r2, r3); r1 ^= r2; r1 = rol7(r1);
c0 = vaddq_u32(c0, c1); c3 ^= c0; c3 = rol16(c3);
c2 = vaddq_u32(c2, c3); c1 ^= c2; c1 = rol12(c1);
c0 = vaddq_u32(c0, c1); c3 ^= c0; c3 = rol8(c3);
c2 = vaddq_u32(c2, c3); c1 ^= c2; c1 = rol7(c1);
vst1q_u8(out + 0, vreinterpretq_u8_u32(vaddq_u32(r0, in0)));
vst1q_u8(out + 16, vreinterpretq_u8_u32(vaddq_u32(r1, in1)));
vst1q_u8(out + 32, vreinterpretq_u8_u32(vaddq_u32(r2, in2)));
vst1q_u8(out + 48, vreinterpretq_u8_u32(vaddq_u32(r3, in3)));
r0 = vaddq_u32(r0, in0);
r1 = vaddq_u32(r1, in1);
r2 = vaddq_u32(r2, in2);
r3 = vaddq_u32(r3, in3);
in3 = vaddq_u32(in3, blkno_inc);
r0 = vaddq_u32(r0, in0);
r1 = vaddq_u32(r1, in1);
r2 = vaddq_u32(r2, in2);
r3 = vaddq_u32(r3, in3);
in3 = vaddq_u32(in3, blkno_inc);