I-Corps: Smart Programming Tools for Improving Software Debugging

这一i-Corps项目的更广泛的影响/商业潜力是，由于采用了新的编译器分析技术，它将显著提高软件可靠性和开发人员的生产率。检测、诊断和修复软件错误非常重要。现有的解决方案是不够的，因为它们经常生成太多的错误警报、遗漏真正的错误、引起令人望而却步的运行时问题或缺乏诊断信息--这些问题对于大型、复杂的软件来说是严重的。这项新技术通过持续分析代码更改并为开发人员提供即时反馈，能够在编程阶段检测和修复复杂的软件错误。这项i-Corps项目旨在评估一种新的编译器分析技术的商业化潜力，并加速将该技术转化为新兴产品和服务。该创新提供了一种增量式分析引擎，其协同地组合了几个基本的静态程序分析，以降低与代码改变成比例的过程间程序分析的平均计算复杂性，而不是整个代码库。凭借快速检测复杂软件错误的能力，该技术可以在集成开发环境(IDE)中使用，以帮助开发人员在将错误引入移动应用程序时立即检测和修复错误。它还可以集成到版本控制系统中，以检测拉请求中的种族，提供比人工代码审查者快得多的反馈；它还可以嵌入到并行化编译器中，以帮助自动并行化程序。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

The broader impact/commercial potential of this I-Corps project it to significantly improve software reliability and developer productivity, as a result of a new compiler analysis technology. Detecting, diagnosing, and fixing software bugs are extremely important. Existing solutions are inadequate because they often generate too many false alarms, miss real errors, incur prohibitive runtime issues, or lack diagnostic information - these problems are acute for large, complex software. The new technology enables detecting and fixing complex software bugs in the programming phase by continuously analyzing code changes and providing developers with instant feedback. The technology will lead to more economical, reliable, and secure software.This I-Corps project aims to evaluate the commercialization potential of a new compiler analysis technology and to accelerate the translation of this technology into emerging products and services. The innovation offers an incremental analysis engine that synergistically combines several fundamental static program analyses to reduce the average computational complexity of inter-procedural program analyses proportional to the code changes instead of the entire code base. With the capability to detect complex software bugs with fast speed, the technology could be used in an integrated development environment (IDE) to help developers detect and fix bugs as soon as they are introduced into mobile apps. It may also be integrated into a version control system to detect races in pull requests, providing much faster feedback than human code reviewers; and it also may be embedded into parallelization compilers to help automatically parallelize programs.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.