Practical Symbolic Static Analysis for Effective Bug Detection
Պահպանված է:
| Հրատարակված է: | PQDT - Global (2025) |
|---|---|
| Հիմնական հեղինակ: | |
| Հրապարակվել է: |
ProQuest Dissertations & Theses
|
| Խորագրեր: | |
| Առցանց հասանելիություն: | Citation/Abstract Full Text - PDF Full text outside of ProQuest |
| Ցուցիչներ: |
Չկան պիտակներ, Եղեք առաջինը, ով նշում է այս գրառումը!
|
MARC
| LEADER | 00000nab a2200000uu 4500 | ||
|---|---|---|---|
| 001 | 3266812750 | ||
| 003 | UK-CbPIL | ||
| 020 | |a 9798297653887 | ||
| 035 | |a 3266812750 | ||
| 045 | 2 | |b d20250101 |b d20251231 | |
| 084 | |a 189128 |2 nlm | ||
| 100 | 1 | |a Guo, Yiyuan | |
| 245 | 1 | |a Practical Symbolic Static Analysis for Effective Bug Detection | |
| 260 | |b ProQuest Dissertations & Theses |c 2025 | ||
| 513 | |a Dissertation/Thesis | ||
| 520 | 3 | |a Static analysis is a powerful approach that automatically detects software bugs by analyzing the program behavior without requiring execution. Many static analysis techniques leverage symbolic methods to achieve greater precision, where they represent unknown values using symbolic variables and describe program states with symbolic formulas. When applying the symbolic method to detect bugs in realistic, million-line codebases, we encounter two challenges. First, the inherent limitations of static analysis often lead to the introduction of many under-constrained symbolic variables, causing significant imprecision. Second, the scalability of symbolic methods is influenced by their memory model, especially when intricate interactions between the memory and numeric values exist or when the analysis is path-sensitive. This thesis presents contributions to improve the precision and scalability of the symbolic method for enabling practical static bug detection at an industrial scale.The first work focuses on the static detection of divide-by-zero bugs, a client where the presence of under-constrained variables severely degrades the precision of symbolic methods. Based on an extensive empirical study, we propose to gather extra evidence about the under-constrained variables by inferring the programmers’ beliefs, which helps the static analyzer to significantly improve its precision by reporting bugs that are more likely to be true.The second work investigates the problem of static buffer overflow detection, which involves mutually dependent program properties of heap and numeric value. Existing methods either sacrifice precision or efficiency in addressing the mutual dependency. To address this dilemma, we propose a novel memory model based on heap disjointness and design a summary-based analysis algorithm, thereby enabling the precise and scalable detection of buffer overflow bugs.The third work concentrates on the scalability problem faced by the path-sensitive memory model in data dependence analysis. We observe that a key performance bottleneck of the analysis is to infer the condition under which storing to a memory location may overwrite its old containing value. Our solution improves the efficiency by decomposing the analysis efforts into stages: We handle most of the memory updates efficiently based on a must-kill relation among the heap stores, while reserving the expensive path-sensitive analysis for the rest. | |
| 653 | |a Motivation | ||
| 653 | |a Programming languages | ||
| 653 | |a Software upgrading | ||
| 653 | |a Open source software | ||
| 653 | |a Linux | ||
| 653 | |a Computer science | ||
| 773 | 0 | |t PQDT - Global |g (2025) | |
| 786 | 0 | |d ProQuest |t ProQuest Dissertations & Theses Global | |
| 856 | 4 | 1 | |3 Citation/Abstract |u https://www.proquest.com/docview/3266812750/abstract/embedded/H09TXR3UUZB2ISDL?source=fedsrch |
| 856 | 4 | 0 | |3 Full Text - PDF |u https://www.proquest.com/docview/3266812750/fulltextPDF/embedded/H09TXR3UUZB2ISDL?source=fedsrch |
| 856 | 4 | 0 | |3 Full text outside of ProQuest |u https://doi.org/10.14711/thesis-hdl151260 |