I'm currently using AVX2 and I have the following problem:
After doing some AVX instructions I have to extract all values and put them into an array, the problem is that the way I found to do it isn't efficient at all...
I did the following things:
uint32_t temp[];
__m256i x, y;
temp[0] = _mm256_extract_epi32(x, 0);
temp[1] = _mm256_extract_epi32(x, 1);
temp[2] = _mm256_extract_epi32(x, 2);
temp[3] = _mm256_extract_epi32(x, 3);
temp[4] = _mm256_extract_epi32(x, 4);
temp[5] = _mm256_extract_epi32(x, 5);
temp[6] = _mm256_extract_epi32(x, 6);
temp[7] = _mm256_extract_epi32(x, 7);
Is there a way to do it better ?
Thanks.