Global data flow analysis of syntax tree intermediate code

While software developers make every effort to develop correct, easily maintainable and efficient programs, it is always possible to make improvements. These improvements may be made to the program code or to the data which is manipulated by the program. The process of measuring the efficiency of program code can be performed by the use of the many tools which are available today, such as performance analyzers. The processs of detecting inefficiency in data manipulation is a more complex task but can nevertheless be represented as the result of data flow analysis. This is analogous to program flow analysis. Performing a data flow analysis on a computer program yields data which can be of use in many areas. It can be considered as bringing relevant information from each part of a program to every other part. It becomes possible to make certain optimizations on the code representation. Decisions as to which types of optimizations are best will be made as a result of consulting the data flow analysis results. Tools can also be constructed which will provide the programmer with information about the state of data which is used by the program. The issue of performing Global Data Flow Analysis on a particular intermediate code representation of a computer program are discussed in this thesis