Synthesizing Performance Constraints for Evaluating and Improving Code Efficiency

TL;DR

This paper introduces WEDGE, a framework that synthesizes performance constraints to generate stress tests for code, enabling better evaluation and optimization of code efficiency, and releases PERFFORGE as a benchmark suite.

Contribution

WEDGE is a novel framework that synthesizes explicit performance constraints to generate targeted stress tests for code, improving evaluation and optimization processes.

Findings

WEDGE significantly slows down code compared to existing tests.

Integrating WEDGE-generated tests enhances code optimization approaches.

PERFFORGE provides a new benchmark for evaluating code efficiency.

Abstract

Large Language Models (LLMs) have been increasingly used to optimize code efficiency. Evaluating their effectiveness and further suggesting optimization opportunities often rely on high-quality tests to demonstrate the performance bottlenecks presented in the program. However, existing approaches rely on a limited set of hand-curated inputs or LLM-generated uninteresting length-stressing tests, failing to reveal more nuanced optimization opportunities. We present WEDGE, a framework for generating performance-stressing input given the program under test. WEDGE synthesizes explicit performance-characterizing constraints in the form of branch conditions to partition the programs' execution space into performance-specific regions. When integrated with the coverage-guided fuzzer, reaching different regions introduces explicit rewards for test generation to explore inefficient implementations. Our evaluation shows that WEDGE introduces a significant slowdown compared to the tests in CodeContests and those claimed to be optimized by existing approaches. From the utility perspective, integrating our tests substantially improves the existing code optimization approaches that rely on test-driven execution feedback. We release PERFFORGE, the performance tests generated by WEDGE, to benchmark future approaches for efficient code generation at https://github.com/UChiSeclab/perfforge.