Third in the "An introduction to compiling for the Cell Broadband Engine™ architecture" series, this tutorial discusses the compiler issues in optimizing code to run efficiently on SIMD-capable processors. In particular, it shows how to optimize code that must run both on the VMX SIMD engine of the PowerPC® core of the Cell Broadband Engine (Cell BE) processor, and also on the SIMD-only Synergistic Processor Elements (SPEs).
In this tutorial
Automatic simdization
Simdization technique highlights
SIMD code generation
Managing complexity
How to support the cross product of all of these?
Example 1: Basic-block and loop-level aggregation
Example 2: Data-size conversion and misalignment
Prerequisites
This tutorial presupposes familiarity with the material in the first tutorial in the series, specifically computer architecture in general and the Cell BE architecture in particular, and basic familiarity with what compilers do. Familiarity with the second tutorial, "Optimizing for the SPE," is also helpful.
