Combining static and dynamic data in code visualization

  • Authors:

  • David Eng

  • Affiliations:

  • McGill University, Montréal, Québec, Canada

  • Venue:
  • Proceedings of the 2002 ACM SIGPLAN-SIGSOFT workshop on Program analysis for software tools and engineering
  • Year:
  • 2002

Quantified Score

Hi-index 0.01

Visualization

Abstract

The task of developing, tuning, and debugging compiler optimizations is a difficult one which can be facilitated by software visualization. There are many characteristics of the code which must be considered when studying the kinds of optimizations which can be performed. Both static data collected at compile-time and dynamic runtime data can reveal opportunities for optimization and affect code transformations. In order to expose the behavior of such complex systems, visualizations should include as much information as possible and accommodate the different sources from which this information is acquired.This paper presents a visualization framework designed to address these issues. The framework is based on a new, extensible language called JIL which provides a common format for encapsulating intermediate representations and associating them with compile-time and runtime data. We present new contributions which extend existing compiler and profiling frameworks, allowing them to export the intermediate languages, analysis results, and code metadata they collect as JIL documents. Visualization interfaces can then combine the JIL data from separate tools, exposing both static and dynamic characteristics of the underlying code. We present such an interface in the form of a new web-based visualizer, allowing JIL documents to be visualized online in a portable, customizable interface.