A C/C++ Smart Contract Vulnerability Detection Method Based on Intermediate Value Transformation

Abstract

Smart contracts have brought new developments to Blockchain technology, quickly finding widespread application in finance, law, and business sectors upon their emergence. However, security vulnerabilities in smart contracts have limited their growth. The existing methods for detecting vulnerabilities in smart contracts primarily focus on Ethereum's Solidity, lacking of vulnerability detection tools based on other smart contract languages such as C/C++. Therefore, this paper investigates the current state of vulnerability detection research for smart contract languages other than Ethereum's Solidity and proposes a vulnerability detection method for C/C++ smart contracts based on intermediate value transformation. This method combines Clang technology with the symbolic execution tool KLEE[l], utilizing the Intermediate Representation (IR) from LLVM (Low Level Virtual Machine) for research, andproduces a detection method capable of identifying eight types of code level vulnerabilities. Finally, it summarizes the current shortcomings of smart contract vulnerability detection methods and future research directions.