Date of Award
Spring 2015
Project Type
Thesis
Program or Major
Computer Science
Degree Name
Master of Science
First Advisor
Phil Hatcher
Second Advisor
Michel Charpentier
Third Advisor
Wheeler Ruml
Abstract
Concurrent programs are notoriously difficult to develop due to the non-deterministic nature of thread scheduling. It is desirable to have a programming language to make such development easier. Tscript comprises such a system. Tscript is an extension of JavaScript that provides multithreading support along with intent specification. These intents allow a programmer to specify how parts of the program interact in a multithreaded context. However, enforcing intents requires run-time memory checks which can be inefficient. This thesis implements an optimization in the Tscript compiler that seeks to improve this inefficiency through static analysis. Our approach utilizes both type inference and dataflow analysis to eliminate unnecessary run-time checks.
Recommended Citation
Hebert, Chris, "Inferring Types to Eliminate Ownership Checks in an Intentional JavaScript Compiler" (2015). Master's Theses and Capstones. 1021.
https://scholars.unh.edu/thesis/1021