CSR: Small: Repurposing Spatial Memory Safety Support in Commodity Processors for Temporal Memory Safety, Other Program Analyses, Hardware-Accelerated Data Structures, and More

CSR：小：重新利用商品处理器中的空间内存安全支持，以实现临时内存安全、其他程序分析、硬件加速数据结构等

• 批准号: 1814430
• 负责人: Dongyoon Lee
• 金额: $ 50万
• 依托单位: Virginia Polytechnic Institute and State University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2018
• 资助国家: 美国
• 起止时间: 2018-07-01 至 2020-05-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1814430&HistoricalAwards=false
• 关键词: CSR; Small; Repurposing; Spatial; Memory

In every generation of computer hardware, researchers and practitioners have found uses for hardware features that their designers did not anticipate. The project aims to find new ways to adapt special-purpose hardware for general and multiple ends, using Intel's Memory Protection Extensions (MPX) as a proof of concept. Intel's designers intended MPX to be used to improve software security by ensuring an important property called spatial memory safety. The project investigates many other surprising uses for MPX.The project explores the idea of viewing special-purpose "hardware" as general-purpose hardware-accelerated "software". In particular, the project aims to repurpose MPX for three broader uses, beyond its original specialized hardware support for spatial memory safety. First, a technique to transparently augment temporal memory safety on top of MPX will be studied to ensure full memory safety. Second, a novel dynamic program analysis framework, which repurposes the MPX as a general-purpose metadata store, will be designed and developed. Last, the project will investigate the potential of MPX as a general-purpose hardware-accelerated data structure.The project will impact both current and future researchers, practitioners, and students. The project will unlock the full potential of Intel's special-purpose hardware, explore the limitations of its instruction set architecture, and thus help design next-generation processors powered by more general-purpose features. These next-generation processors will enable more secure, reliable, and performant software. New undergraduate course modules will be developed to integrate research discovery with teaching. A K-12 summer class will be offered as an outreach and diversity program. The project also includes a plan to offer a program analysis workshop for non-computer scientists to support the broader research community.The project's main results will be software prototypes and hardware simulations, along with benchmarks, documentation, and research papers. The source code, benchmark applications, and documentation will be made publicly available on GitHub. The papers will be available on the dedicated project website as well. The project repository will be maintained for at least five years beyond the end of the project. All of these resources will be available from the project web page: http://research.cs.vt.edu/gpx.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.

在每一代计算机硬件中，研究人员和实践者都发现了设计者没有预料到的硬件功能的用途。 该项目旨在寻找新的方法来适应通用和多用途硬件，使用英特尔的内存保护扩展（MPX）作为概念验证。英特尔的设计者打算使用MPX来提高软件安全性，确保一个重要的属性，称为空间内存安全。该项目调查了MPX的许多其他令人惊讶的用途。该项目探索了将专用“硬件”视为通用硬件加速“软件”的想法。特别是，该项目旨在将MPX重新用于三个更广泛的用途，超越其最初的空间内存安全专用硬件支持。首先，将研究在MPX之上透明地增强时态存储器安全的技术，以确保完全的存储器安全。第二，一个新的动态程序分析框架，它重新利用MPX作为一个通用的元数据存储，将设计和开发。最后，本计画将探讨MPX作为一种通用硬体加速资料结构的潜力，并将影响目前及未来的研究人员、从业者及学生。该项目将释放英特尔专用硬件的全部潜力，探索其指令集架构的局限性，从而帮助设计由更多通用功能驱动的下一代处理器。这些下一代处理器将实现更安全、可靠和高性能的软件。将开发新的本科课程模块，将研究发现与教学相结合。K-12暑期班将作为一个推广和多样性计划提供。该项目还计划为非计算机科学家提供程序分析研讨会，以支持更广泛的研究社区。该项目的主要成果将是软件原型和硬件模拟，沿着基准测试，文档和研究论文。源代码、基准应用程序和文档将在GitHub上公开。这些文件也将在专门的项目网站上提供。项目资料库将在项目结束后至少维持五年。所有这些资源都可以从项目网页上获得：http://research.cs.vt.edu/gpx.This奖项反映了NSF的法定使命，并被认为值得通过使用基金会的知识价值和更广泛的影响审查标准进行评估来支持。

项目成果

PeX: A Permission Check Analysis Framework for Linux Kernel

• 发表时间: 2019
• 作者: Tong Zhang;Wenbo Shen;Dongyoon Lee;Changhee Jung;Ahmed M. Azab;Ruowen Wang

Why aren’t regular expressions a lingua franca? an empirical study on the re-use and portability of regular expressions

• DOI: 10.1145/3338906.3338909
• 发表时间: 2019-08
• 期刊: Proceedings of the 2019 27th ACM Joint Meeting on European Software Engineering Conference and Symposium on the Foundations of Software Engineering
• 作者: James C. Davis;IV LouisG.Michael;Christy A. Coghlan;Francisco Servant;Dongyoon Lee

BOGO: Buy Spatial Memory Safety, Get Temporal Memory Safety (Almost) Free

• DOI: 10.1145/3297858.3304017
• 发表时间: 2019-04
• 期刊: Proceedings of the Twenty-Fourth International Conference on Architectural Support for Programming Languages and Operating Systems
• 作者: T. Zhang;Dongyoon Lee;Changhee Jung

The impact of regular expression denial of service (ReDoS) in practice: an empirical study at the ecosystem scale

• DOI: 10.1145/3236024.3236027
• 发表时间: 2018-10
• 期刊: Proceedings of the 2018 26th ACM Joint Meeting on European Software Engineering Conference and Symposium on the Foundations of Software Engineering
• 作者: James C. Davis;Christy A. Coghlan;Francisco Servant;Dongyoon Lee

Compiler-directed soft error resilience for lightweight GPU register file protection

• DOI: 10.1145/3385412.3386033
• 发表时间: 2020-06
• 期刊: Proceedings of the 41st ACM SIGPLAN Conference on Programming Language Design and Implementation
• 作者: Hongjune Kim;Jianping Zeng;Qingrui Liu;Mohammad Abdel-Majeed;Jaejin Lee;Changhee Jung