Previous
The performances
Parent
Optimisation of a reduction
Outline
Next
The reduction_vectorize.h
Chapter 8.3 : The vectorization of reduction
8.3.1) The reduction_vectorize.h
8.3.2) The reduction_vectorize.cpp
8.3.3) The main_vectorize.cpp
8.3.4) The CMakeLists.txt
We, again, have to inform the compiler :
The tables (or pointers) are aligned with
__builtin_assume_aligned
The tables (or pointers) do not overhead eachother with
__restrict__
Enable the vectorization with
-O3 -ftree-vectorize -march=native -mtune=native -mavx2
Previous
The performances
Parent
Optimisation of a reduction
Outline
Next
The reduction_vectorize.h