DEPARTMENT OF COMPUTER SCIENCE
DOCTORAL DISSERTATION DEFENSE


Candidate: YOUNG G. PARK
Advisor: BENJAMIN GOLDBERG

SEMANTIC PROGRAM ANALYSES FOR STORAGE MANAGEMENT OPTIMIZATIONS IN FUNCTIONAL LANGUAGE IMPLEMENTATIONS

11:00 a.m., Friday, September 27, 1991
12th floor, 719 Broadway



Abstract

One of the major overheads in implementing functional languages in both uniprocessor and multiprocessor environments is the storage management overhead due to dynamic allocation and automatic reclamation of indefinite- extent storage. We investigate compiler optimization to reduce such overhead by statically inferring the lifetime information about dynamically-allocated objects.

We have developed a set of compile-time semantic analyses for a higher-order monomorphic strict functional language based on denotational semantics and abstract interpretation:

Those analyses are extended to both polymorphic and non-strict (either normal- order evaluation or lazy evaluation) languages.

Using statically inferred escape information, we have proposed a variety of storage management optimization techniques including stack allocation, explicit reclamation, in-place reuse, reference counting elimination, block allocation/reclamation, and improving generational garbage collection.