#include static inline void gfmulby2(__m128i a,__m128i* res) { *res = _mm_srai_epi32(a,31); *res = _mm_shuffle_epi32(*res,0x57); *res = _mm_and_si128(*res,_mm_set_epi32(0x00,0x01,0x00,0x87)); *res = _mm_xor_si128(*res,_mm_slli_epi64(a,0x01)); }