VectoriseD maTh C++ Library for Fast Math
copied from cf-staging / vdtVDT is a library of mathematical functions, implemented in double and single precision. The implementation is fast and with the aid of modern compilers (e.g. gcc 4.7) autovectorisable.