Frank Mori Hess a écrit :
Isn't SIMD-type stuff what valarray was designed for? I don't think so or you'll have to show me where in the STL files of valarray this SIMD implementation lives. I don't think a SIMD optimized valarray requires its own DSEL, there is the macstl implementation that has been around for a while. It's not SIMD valarray, vec<T> acts as SIMD POD values. I know about macstl (is it still maintained btw ?) as I've designed the very same kind of library during my PHD but vec<T> as a complete different use case aka writing low level SIMD enabled numerical algorthims without necessarily working around an array abstraction.
-- ___________________________________________ Joel Falcou - Assistant Professor PARALL Team - LRI - Universite Paris Sud XI Tel : (+33)1 69 15 66 35