High-Performance Recipes for IA-32 Platforms
by Richard Gerber, Aart J.C. Bik, Kevin B. Smith and Xinmin Tian

More about this book

Excerpt Description
Vectorization diagnostics provide useful information on places where automatic vectorization has been successful and where it has failed. This information can subsequently be used to make code more amenable to vectorization or to disable vectorization of fragments where SIMD technology adversely affects performance. This excerpt illustrates this process.