Infrastructure for the Generation of Functional Data-Flow Views for Automotive Embedded Systems

Data-flow visualization for source code can help software developers and software architects to understand code graphically. In this thesis, an infrastructure for dataflow visualization is created to analyze the C source code of an embedded system of a truck. Several commercial and open-source tools for data-flow analysis are investigated and a definition for data-flow is found. A data-flow analysis tool chain consisting of FLex, a lexical analyzer generator, Bison, a parser generator, and a hand-written data-flow analysis is implemented. The tool chain saves data-flow information from the source code into an intermediate representation which can be used to create visualizations. Software developers and architects are interviewed to gather information about how data-flow visualizations are used at Scania and how the tool chain can be improved