YoVDO

Unleashing Mayhem on Binary Code

Offered By: IEEE via YouTube

Tags

Exploit Development Courses Cybersecurity Courses Symbolic Execution Courses Vulnerability Research Courses

Course Description

Overview

Explore the intricacies of automatic exploit generation in binary programs through this IEEE Symposium on Security & Privacy presentation. Delve into the innovative Mayhem system, designed to uncover exploitable bugs and generate working shell-spawning exploits in executable programs without debugging information. Learn about two groundbreaking techniques: hybrid symbolic execution and index-based memory modeling, which address challenges in managing execution paths and reasoning about symbolic memory indices. Discover how Mayhem successfully identified 29 exploitable vulnerabilities across Linux and Windows programs, including two previously undocumented issues. Gain insights into the system's architecture, symbolic execution process, safety policies, and optimization techniques for efficient vulnerability detection and exploit generation in binary code.

Syllabus

Intro
Automatic Exploit Generation Challenge
Ghostscript v8.62 Bug
Generating Exploits
Unleashing Mayhem
How Mayhem Works: Symbolic Execution
Path Predicate = II
Safety Policy in Mayhem
Challenges
Current Resource Management in Symbolic Execution
Offline Execution
Online Execution
Mayhem: Hybrid Execution
Symbolic Indices
Another Cause: Table Lookups
Method 1: Concretization
Method 2: Fully Symbolic
Step 1 - Find Bounds
Step 2 - Index Search Tree Construction
Fully Symbolic vs. Index-based Memory Modeling Time
Index Search Tree Optimization: Piecewise Linear Approximation
Conclusion


Taught by

IEEE Symposium on Security and Privacy

Tags

Related Courses

Formal Software Verification
University System of Maryland via edX
Software Analysis & Testing
Georgia Institute of Technology via Udacity
Computer Systems Security
Massachusetts Institute of Technology via MIT OpenCourseWare
Reverse Engineering 3201: Symbolic Analysis
OpenSecurityTraining2 via Independent
angr: Binary Analysis Framework - Demonstration and Analysis
New York University (NYU) via YouTube