3. Reduction of Waiting Time for Calculation

When CPU performance analysis reports show higher ratio of waiting time for floating-point/integer calculation on cycle accounting, there might be room to check and promote compiler optimizations. Effective techniques for such cases are explained in this section.

• 3.1. Fission of Loop with Large Loop Body
• 3.2. Striping of Innermost Loop with Small Iteration Count

On cycle accounting, waiting time for calculation and cache access is distinguished for each CPU cycle by execution status of the oldest in-flight CPU instruction in the program order. Therefore, techniques for Reduction of Waiting Time for Cache Access might lead to reduction of waiting time for calculation as a result, for example in case of a program where many calculations are chained after an array reference within a loop.