Detecting Memory-Related Bugs by Tracking Heap Memory Management of C++ Smart Pointers

Ma, Xutong; Yan, Jiwei; Wang, Wei; Yan, Jun; Zhang, Jian; Qiu, Zongyan

doi:10.1109/ase51524.2021.9678836

Cited by 7 publications

(1 citation statement)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Test cases do not need to be artificially designed, which solves the software testing limitation caused by incomplete test cases. However, this method also has the disadvantages of slow detection speed and severe false positives [ 2 ]. Therefore, reducing the False Positive Rate of memory leak detection to improve the detection efficiency is the focus of this paper.…”

Section: Introductionmentioning

confidence: 99%

MLD: An Intelligent Memory Leak Detection Scheme Based on Defect Modes in Software

Yuan

Zhou

Pan

et al. 2022

Entropy

View full text Add to dashboard Cite

With the expansion of the scale and complexity of multimedia software, the detection of software defects has become a research hotspot. Because of the large scale of the existing software code, the efficiency and accuracy of the existing software defect detection algorithms are relatively low. We propose an intelligent memory leak detection scheme MLD based on defect modes in software. Based on the analysis of existing memory leak defect modes, we summarize memory operation behaviors (allocation, release and transfer) and present a state machine model. We employ a fuzzy matching algorithm based on regular expression to determine the memory operation behaviors and then analyze the change in the state machine to assess the vulnerability in the source code. To improve the efficiency of detection and solve the problem of repeated detection at the function call point, we propose a function summary method for memory operation behaviors. The experimental results demonstrate that the method we proposed has high detection speed and accuracy. The algorithm we proposed can identify the defects of the software, reduce the risk of being attacked to ensure safe operation.

show abstract

Section: Introductionmentioning

confidence: 99%