Genetic Improvement in the Shackleton Framework for Optimizing LLVM Pass Sequences

TL;DR

This paper introduces a genetic improvement approach within the Shackleton framework to automatically discover LLVM pass sequences that optimize runtime performance for specific problems, surpassing default optimization levels.

Contribution

It presents a novel method using genetic improvement to automatically generate problem-specific LLVM pass sequences within the Shackleton framework.

Findings

Achieved an average of 3.7% runtime improvement over -O3 optimization.

Developed a pass-level patch representation for genetic programming.

Demonstrated automatic optimization without expert domain knowledge.

Abstract

Genetic improvement is a search technique that aims to improve a given acceptable solution to a problem. In this paper, we present the novel use of genetic improvement to find problem-specific optimized LLVM pass sequences. We develop a pass-level patch representation in the linear genetic programming framework, Shackleton, to evolve the modifications to be applied to the default optimization pass sequences. Our GI-evolved solution has a mean of 3.7% runtime improvement compared to the -O3 optimization level in the default code generation options which optimizes on runtime. The proposed GI method provides an automatic way to find a problem-specific optimization sequence that improves upon a general solution without any expert domain knowledge. In this paper, we discuss the advantages and limitations of the GI feature in the Shackleton Framework and present our results.