| Tesi etd-09032025-101734 | 
    Link copiato negli appunti
  
    Tipo di tesi
  
  
    Tesi di laurea magistrale
  
    Autore
  
  
    GERACI, EDOARDO  
  
    URN
  
  
    etd-09032025-101734
  
    Titolo
  
  
    Target-Oriented Symbolic Execution for Bug Detection in WebAssembly binaries
  
    Dipartimento
  
  
    INGEGNERIA DELL'INFORMAZIONE
  
    Corso di studi
  
  
    COMPUTER ENGINEERING
  
    Relatori
  
  
    relatore Prof. Lettieri, Giuseppe
supervisore Prof. Costa, Gabriele
  
supervisore Prof. Costa, Gabriele
    Parole chiave
  
  - Manticore
- security
- static analysis
- symbolic execution
- target-oriented approach
- vulnerability detection
- Wassail
- WebAssembly (WASM)
    Data inizio appello
  
  
    02/10/2025
  
    Consultabilità
  
  
    Non consultabile
  
    Data di rilascio
  
  
    02/10/2028
  
    Riassunto
  
  During security evaluations, particularly for binary black-box targets, symbolic execution is a valuable technique for security analysts. It enables the verification of assumptions, the discovery of relevant program behaviors, and the identification of points of interest for deeper analysis. Most existing symbolic execution tools adopt a coverage-first approach, seeking to explore as many reachable functions as possible.
The primary limitation of symbolic execution, however, is its high computational cost. Because the tool must emulate program execution and explore all possible symbolic states, it quickly faces the well-known problem of path explosion.
In this thesis, we propose a target-oriented approach to mitigate this challenge. Rather than maximizing coverage, our method concentrates on determining the constraints required to reach specific states of interest, defined through custom rules provided by the analyst. To realize this approach, we extend state-of-the-art tools such as Wassail and Manticore, adapting them to support our target-oriented methodology. In addition, we address performance bottlenecks by introducing a parallelized execution strategy that leverages memoization, significantly improving the speed and practicality of the tool.
The primary limitation of symbolic execution, however, is its high computational cost. Because the tool must emulate program execution and explore all possible symbolic states, it quickly faces the well-known problem of path explosion.
In this thesis, we propose a target-oriented approach to mitigate this challenge. Rather than maximizing coverage, our method concentrates on determining the constraints required to reach specific states of interest, defined through custom rules provided by the analyst. To realize this approach, we extend state-of-the-art tools such as Wassail and Manticore, adapting them to support our target-oriented methodology. In addition, we address performance bottlenecks by introducing a parallelized execution strategy that leverages memoization, significantly improving the speed and practicality of the tool.
    File
  
  | Nome file | Dimensione | 
|---|---|
| La tesi non è consultabile. | |
 
		