Hi, I repeatedly apply IAND (bitwise and) on large integer vectors. I was wondering whether internally the command uses simd to accelerate the result. If not, what compiler directives or flags do I need to use to activate this? Can I use Intel Advanced Vector Extension?
I use intel parallel studio 2016.
-QxHost to take advantage of the simd instructions available on your build machine (if running on same architecture)
-Qopt-report:4 to get the optrpt listing to see if you got vectorization, or possibly why not.