_mm_set1_epi32
__m128i cl128 = _mm_set1_epi32(cl); \
__m128i ch128 = _mm_set1_epi32(ch); \
q[1] ^= _mm_set1_epi32(sk[1]);
q[2] ^= _mm_set1_epi32(sk[2]);
q[3] ^= _mm_set1_epi32(sk[3]);
q[4] ^= _mm_set1_epi32(sk[4]);
q[5] ^= _mm_set1_epi32(sk[5]);
q[6] ^= _mm_set1_epi32(sk[6]);
q[7] ^= _mm_set1_epi32(sk[7]);
y0 = x & _mm_set1_epi32(0x000000FF);
y1 = _mm_slli_epi32(x & _mm_set1_epi32(0x00003F00), 2);
y2 = _mm_srli_epi32(x & _mm_set1_epi32(0x0000C000), 6);
y3 = _mm_slli_epi32(x & _mm_set1_epi32(0x000F0000), 4);
y4 = _mm_srli_epi32(x & _mm_set1_epi32(0x00F00000), 4);
y5 = _mm_slli_epi32(x & _mm_set1_epi32(0x03000000), 6);
y6 = _mm_srli_epi32(x & _mm_set1_epi32(0xFC000000), 2);
q[0] ^= _mm_set1_epi32(sk[0]);
q[0] ^= _mm_set1_epi32(sk[0]);
q[1] ^= _mm_set1_epi32(sk[1]);
q[2] ^= _mm_set1_epi32(sk[2]);
q[3] ^= _mm_set1_epi32(sk[3]);
q[4] ^= _mm_set1_epi32(sk[4]);
q[5] ^= _mm_set1_epi32(sk[5]);
q[6] ^= _mm_set1_epi32(sk[6]);
q[7] ^= _mm_set1_epi32(sk[7]);
y0 = x & _mm_set1_epi32(0x000000FF);
y1 = _mm_srli_epi32(x & _mm_set1_epi32(0x0000FC00), 2);
y2 = _mm_slli_epi32(x & _mm_set1_epi32(0x00000300), 6);
y3 = _mm_srli_epi32(x & _mm_set1_epi32(0x00F00000), 4);
y4 = _mm_slli_epi32(x & _mm_set1_epi32(0x000F0000), 4);
y5 = _mm_srli_epi32(x & _mm_set1_epi32(0xC0000000), 6);
y6 = _mm_slli_epi32(x & _mm_set1_epi32(0x3F000000), 2);
x12 = _mm_add_epi32(_mm_set1_epi32(blkno),
x12 = _mm_add_epi32(_mm_set1_epi32(blkno),