Intel® Advisor Help
Using the Intel® C/C++ and Fortran Compilers, you can add SIMD parallelism via unrolling a loop so that it takes advantage of packed SIMD instructions to perform the same operation on multiple data elements with a single instruction, so that the loop can execute more efficiently.
You can enable automatic vectorization with the Intel® Compiler at optimization levels of O2 and higher.
Many loops are vectorized automatically, but when automatic vectorization does not happen, you can try to vectorize loops by making simple code modifications.
Consider the following methods of loop vectorization:
Hinting the compiler to vectorize, which means changing the source code so that the compiler vectorizes the code automatically.
Or forcing the compiler to vectorize the loop, which means adding pragmas into your source code.