Efficient and precise points-to analysis: modeling the heap by merging equivalent automata

This artifact is provided to enable the results of all two research questions (RQ1 and RQ2) in our companion paper, i.e., the results in Figure 8, Figure 9, Table 1 and Table 2, to be reproduced. In RQ1, we show that MAHJONG is lightweight for large programs and how MAHJONG can alleviate the heap over-partitioning problem suffered by the allocation-site abstraction effectively for type-dependent clients. In RQ2, we show that MAHJONG can significantly accelerate different types of mainstream context-sensitive points-to analyses while achieving nearly the same precision as the allocation-site abstraction for type-dependent clients. The artifact contains MAHJONG and Doop (a state-of-the-art whole-program points-to analysis framework for Java) to reproduce the results. Size of the artifact: 115MB MD5 sum of the artifact: 619a2d64be9fa61b158f2b04ad55e4a0