Efficient and precise modeling of exceptions for the analysis of Java programs

Abstract

The Factored Control Flow Graph, FCFG, is a novel representation of a program's intraprocedural control flow, which is designed to efficiently support the analysis of programs written in languages, such as Java, that have frequently occurring operations whose execution may result in exceptional control flow. The FCFG is more compact than traditional CFG representations for exceptional control flow, yet there is no loss of precision in using the FCFG. In this paper, we introduce the FCFG representation and outline how standard forward and backward data flow analysis algorithms can be adapted to work on this representation. We also present empirical measurements of FCFG sizes for a large number of methods obtained from a variety of Java programs, and compare these sizes with those of a traditional CFG representation.

Representing concerns in source code

“…1 The elicitation is performed by applying the selection criterion to all the elements declared in P , and by applying the analysis function to P for all relation names specified in the mapping function. In practice, this step is performed via standard static analyses, such as parsing [35], type-checking [3], control-or data-flow analysis [84,91], and exception flow analysis [27,111,114,120,125].…”

“…Finally, the mapping function does not support exception handling. Exception handling introduces a particular type of control-flow that can be difficult to abstract [27,111,114,125]. For the purpose of experimenting with concern graphs and validating the thesis, we chose to leave exception handling aside.…”

The Jalapeño virtual machine

“…This section provides an overview of the Jalapeño optimizing compiler; further details are available elsewhere. [15][16][17] The optimizing compiler's structure is shown in Figure 4.…”

“…(Extra care is required when performing data flow analysis or code motion on these extended basic blocks.) 16,17 These intermediate representations also include space for the caching of such optional auxiliary information as reaching-definition 18 sets, dependence graphs, and encodings of loop-nesting structure.…”

“…25 After the MIR is constructed, live variable analysis is performed to determine the live ranges of symbolic registers and the stack variables that hold object references at garbage-collection-safe points. The standard live variable analysis 18 has been modified to handle the extended basic blocks of the factored control flow graph as described by Choi, et al 16 Next, the optimizing compiler employs the linear scan global register-allocation algorithm 26 to assign physical machine registers to symbolic MIR registers. This algorithm is not based on graph coloring, but greedily allocates physical to symbolic registers in a single linear time scan of the symbolic registers' live ranges.…”