2. Promoting Vectorization

Exploiting computational performance of A64FX processors requires well-vectorized object codes generated by compilers. Effective techniques for such objective are explained in this section.

• 2.1. Interchange of Innermost Loop with Data Dependency
• 2.2. Interchange of Innermost Loop with Small Iteration Count
• 2.3. Fission of Imperfectly Nested Loops