This repository contains data for "Exploiting Domain and Program Structure to Synthesize Efficient and Precise Data Flow Analyses" by Elena Sherman and Matthew B. Dwyer
Please let us know if something is unclear.
There are three top-level directories src - the source code of the analysis, it has its own readme inside artifacts -- contains 19 classes used in the experiments domain -- contains 27 files with domain descriptions
Mapping from the domain files ids to their descriptions: dom4 - small BL 1 dom41 - small BL 2 dom42 - small BL 3 dom5 - small DD 1 dom51 - small DD 2 dom52 - small DD 3 dom5_4 - small DD_ 1 dom51_41 - small DD_ 2 dom52_42 - small DD_ 3
dom6 - medium BL 1 dom61 - medium BL 2 dom62 - medium BL 3 dom7 - medium DD 1 dom71 - medium DD 2 dom72 - medium DD 3 dom7_6 - medium DD_ 1 dom71_61 - medium DD_ 2 dom72_62 - medium DD_ 3
dom8 - large BL 1 dom81 - large BL 2 dom82 - large BL 3 dom9 - large DD 1 dom91 - large DD 2 dom92 - large DD 3 dom9_8 - large DD_ 1 dom91_81 - large DD_ 2 dom92_82 - large DD_ 3
experiment -- contains two folders accuracy and time with the corresponding experiment data experiment/accuracy -- contains the results of analyses as state description in smt2 format. Each file contains data for all methods for corresponding domain. sN suffix means no black transfer function and sY means that block transfer function was enabled. experiment/time -- timeData1 contains running time(x3) per method for small domains, timeData2 for medium domains and timeData3 for large domains