Static Analysis techniques for inferring program properties
Speaker: Abu Naser Masud
Abu Naser Masud is a researcher in the School of Innovation, Design, and Engineering (IDT) at Malardalen University, Sweden. He is an alumnus of the Technical University of Madrid where he graduated with cum laude in 2013 in Computer Science. He completed his MSc in computational logic from the International Center for Computational Logic (ICCL) at University of Technology Dresden, Germany. After completing his PhD, he joined as a post doc at MDH.
His research interest is on analysis and verification of software systems, WCET analysis, termination analysis, symbolic computation, etc., and has published several international journal and peer-reviewed conference papers on these topics. Currently, he is involved in the SPACES project (Static Program analysis for Complex Embedded Systems) aiming at designing static program analyzer for multithreaded C code where MDH, Ericsson, and Evidente are involved as project partners.
He was a former assistant professor in computer science and engineering discipline at Khulna University, Bangladesh.
Static program analyses infer runtime program properties approximately without actually running the program. These properties have a range of usefulness: from deciding the quality of the code to detecting bugs for example. Sometimes these properties can be quantitative (e.g. computing the worst-case execution time (WCET)), or non-quantitative (e.g. does the code terminate?).
In this talk, I shall present an overview of my research. I shall talk very briefly on the following topics: (i) symbolic cost analysis of imperative programs (ii) termination analysis (iii) demand driven program slicing (iv) data race detection of multi-threaded programs.