Chapter 8: Memory Usage Improvement Using Runtime Alias Detection.

Region-based memory management replaces runtime garbage collection and it enables each memory operation to be constant time operation. This is very important feature for real time applications. However, there are some kinds of programs which are not amenable to region inference. When executed on region-based systems, these programs can cause significant memory leakage and in the worst case, they cannot finish their execution because of memory shortage. In this paper, we present a technique to improve memory usage of Tofte/- Talpin region-based system[8] . Our technique adds some changes to Storage Mode Analysis (SMA)[2], which is a succeeding phase of region inference, and delays some decisions till runtime as to whether or not it is possible to overwrite existing objects. Our method is especially useful for a program compiled separately, where we cannot see the contexts in which top-level functions are called. We implemented this technique to MLKit[7][3] and confirmed that the amount of memory used during execution is reduced for some programs. [ABSTRACT FROM AUTHOR]