Conferences and supporting programme
How to Find Concurrency Issues in C and C++
Most software today uses multi-core processors, so that various tasks of the program run in parallel. This is known as multithreading. Multithreading can speed up your program. But the threads will need to communicate with each other and synchronize their work. So, it can also lead to concurrency issues, such as deadlock and data race. And concurrency issues lead to serious security problems. SEI CERT C/C++ and CWE provide guidelines for secure coding to avoid issues like these. Both include a significant number of items dealing with correct implementation of concurrency. Developers often use parallel debuggers to analyze these issues. But the debugging environment differs from the actual execution environment. So, developers can’t reproduce these issues. There’s an easier, scalable way to find concurrency issues. In this session, we’ll show you how to find these issues by using static code analysis . You’ll learn how it: • Creates a thread summary for each thread encountered in a translation unit. • Combines all thread summaries to perform an in-depth analysis of every type of concurrency issue in the complete application.
--- Date: 27.02.2019 Time: 4:30 PM - 5:00 PM Location: Conference Counter NCC Ost