void _SIMD_cmpge_pd(__SIMDd a, __SIMDd b, void** resultPtr)
{
__SIMDd* result = (__SIMDd*)malloc(sizeof(__SIMDd));
*resultPtr = result;
#ifdef USE_SSE
*result = _mm_cmpge_pd(a,b);
#elif defined USE_AVX
*result = _mm256_cmp(a,b,29);
#elif defined USE_IBM
*result = vec_cmpge(a,b);
#endif
}
inline vector4db operator<(const vector4d& lhs, const vector4d& rhs)
{
return _mm256_cmp(lhs, rhs, _CMP_LT_OQ);
}