Loop splitting

Loop splitting is a compiler optimization technique. It attempts to simplify a loop or eliminate dependencies by breaking it into multiple loops which have the same bodies but iterate over different contiguous portions of the index range.

Loop peeling is a special case of loop splitting which splits any problematic first (or last) few iterations from the loop and performs them outside of the loop body.

Suppose a loop was written like this:

 int p = 10;
 for (int i=0; i<10; ++i)
 {
   y[i] = x[i] + x[p];
   p = i;
 }

Notice that p = 10 only for the first iteration, and for all other iterations, p = i - 1. A compiler can take advantage of this by unwinding (or "peeling") the first iteration from the loop.

After peeling the first iteration, the code would look like this:

 y[0] = x[0] + x[10];
 for (int i=1; i<10; ++i)
 {
   y[i] = x[i] + x[i-1];
 }

This equivalent form eliminates the need for the variable p inside the loop body.

Loop peeling was introduced in gcc in version 3.4. More generalised loop splitting was added in GCC 7.^[1]

Apparently the term "peeling" was for the first time used by Cannings, Thompson and Skolnick^[2] in their 1976 paper on computational models for (human) inheritance. There the term was used to denote a method for collapsing phenotypic information onto parents. From there the term was used again in their papers, including their seminal paper on probability functions on complex pedigrees.^[3]

In compiler technology, the term first turned up in late 1980s papers on VLIW and superscalar compilation, including ^[4] and.^[5]

• Kennedy, Ken; Allen, Randy (2002). "Chapter 5.7. Index-Set Splitting - Chapter 5.7.2. Loop Peeling". Optimizing Compilers for Modern Architectures: A Dependence-Based Approach (2011 digital print of 1st ed.). Academic Press / Morgan Kaufmann Publishers / Elsevier. pp. 211–212. ISBN 978-1-55860-286-1. LCCN 2001092381.