SHF:Small: Debug Information Validation for Optimizing Compilers

Georgia Tech Research Corporation, Atlanta GA

Investigators

Abstract

Almost all modern production software is compiled with optimization. Debugging optimized code is an important part of the software-development process. Perhaps less known to application developers is that production compilers do contain bugs, leading to incorrect debug information. Wrong debug information causes debuggers to either crash or display wrong variable values. This project will devise novel, effective methodologies, techniques, and tools to validate the debug information generation in complex, real-world optimizing compilers. The project's novelties are (1) providing a deeper and more thorough understanding of the fundamental and practical challenges for validating debug information for optimizing compilers, and (2) addressing the two most fundamental aspects of debug-information validation: test-program generation and debug-action generation. The project's impacts are (1) increasing the capability of developers to debug optimized code and deployed software, and (2) more reliable and usable debuggers thereby indirectly improving the quality of software. The goal of this project is to explore novel, practical techniques for validating debug information generated by optimizing compilers. It focuses on developing a systematic framework to enhance the correctness of debug information associated with optimized code. This project explores three main directions: (1) developing new strategies to generate actionable programs; (2) emitting comprehensive debug sequences for actionable programs; and (3) generalizing techniques for different programming languages. The project significantly advances the state-of-the-art in debug information testing and validation. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.