• Login
    View Item 
    •   SMARTech Home
    • Georgia Tech Theses and Dissertations
    • Georgia Tech Theses and Dissertations
    • View Item
    •   SMARTech Home
    • Georgia Tech Theses and Dissertations
    • Georgia Tech Theses and Dissertations
    • View Item
    JavaScript is disabled for your browser. Some features of this site may not work without it.

    Hardware-Assisted Processor Tracing for Automated Bug Finding and Exploit Prevention

    Thumbnail
    View/Open
    YAGEMANN-DISSERTATION-2022.pdf (1.171Mb)
    Date
    2022-05-10
    Author
    Yagemann, Carter
    Metadata
    Show full item record
    Abstract
    The proliferation of binary-only program analysis techniques like fuzz testing and symbolic analysis have lead to an acceleration in the number of publicly disclosed vulnerabilities. Unfortunately, while bug finding has benefited from recent advances in automation and a decreasing barrier to entry, bug remediation has received less attention. Consequently, analysts are publicly disclosing bugs faster than developers and system administrators can mitigate them. Hardware-supported processor tracing within commodity processors opens new doors to observing low-level behaviors with efficiency, transparency, and integrity that can close this automation gap. Unfortunately, several trade-offs in its design raise serious technical challenges that have limited widespread adoption. Specifically, modern processor traces only capture control flow behavior, yield high volumes of data that can incur overhead to sift through, and generally introduce a semantic gap between low-level behavior and security relevant events. To solve the above challenges, I propose control-oriented record and replay, which combines concrete traces with symbolic analysis to uncover vulnerabilities and exploits. To demonstrate the efficacy and versatility of my approach, I first present a system called ARCUS, which is capable of analyzing processor traces flagged by host-based monitors to detect, localize, and provide preliminary patches to developers for memory corruption vulnerabilities. ARCUS has detected 27 previously known vulnerabilities alongside 4 novel cases, leading to the issuance of several advisories and official developer patches. Next, I present MARSARA, a system that protects the integrity of execution unit partitioning in data provenance-based forensic analysis. MARSARA prevents several expertly crafted exploits from corrupting partitioned provenance graphs while incurring little overhead compared to prior work. Finally, I present Bunkerbuster, which extends the ideas from ARCUS and MARSARA into a system capable of proactively hunting for bugs across multiple end-hosts simultaneously, resulting in the discovery and patching of 4 more novel bugs.
    URI
    http://hdl.handle.net/1853/67192
    Collections
    • College of Computing Theses and Dissertations [1191]
    • Georgia Tech Theses and Dissertations [23877]

    Browse

    All of SMARTechCommunities & CollectionsDatesAuthorsTitlesSubjectsTypesThis CollectionDatesAuthorsTitlesSubjectsTypes

    My SMARTech

    Login

    Statistics

    View Usage StatisticsView Google Analytics Statistics
    facebook instagram twitter youtube
    • My Account
    • Contact us
    • Directory
    • Campus Map
    • Support/Give
    • Library Accessibility
      • About SMARTech
      • SMARTech Terms of Use
    Georgia Tech Library266 4th Street NW, Atlanta, GA 30332
    404.894.4500
    • Emergency Information
    • Legal and Privacy Information
    • Human Trafficking Notice
    • Accessibility
    • Accountability
    • Accreditation
    • Employment
    © 2020 Georgia Institute of Technology