/ ~~~~~ __ _. Ab stract An efficient system to perform garbage collection in para ~ lei with list operations is proposed and its correctness is proven. The system consists of two independent processes sharing a common memory. One process is performed by the list process or (LP) for list processing and the other by the garbage collector (GC) for marking active nodes and collecting garbage nodes. The system is derived by using ~ ih the correctness and effic iency arguments. Assuming that memory references are Indivisible the system sa tisfies the following properties:-No critical sections are needed in the entire system.-The time to perform the marking phase by the GC is independent of the size of memory, but depends Onl y on the number of active nodes.-Nodes on the free list need not be marked during the marking phase by the GC.-Minimum overheads are introduced to the LP.-Only two extra bits for encoding four colors are needed for each node. Efficiency results show that the parallel system is usually significantly more efficien t in terms of storage and time than the sequential stack algorithm. , IS t sv ~~~~~~~~ • ~~-; ~ ~~ In this paper we propose an efficient system to perform garbage collection in parallel with list operations, and prove the correctness of the system. The system consists of two Independent processes sharing a common memory. One process Is perf ormed by the list processor (LP) for list processing and the other by the garbage collector (GC) f or marking active nodes and collecting garbage nodes. The system is det ived by using ~~~ the correctness and efficiency arguments (see A ppendix I). Assuming that memory references are indivisible (see Section 4.1), the system sat isfies the following properties: P1. No critical sections are needed in the entire system. P2. The time to perform the marking phase by the GC is independent of the size of memory, but depends only on the number of active nodes. P3. Nodes on the free list need not be marked during the marking phase by the GC. P4. Minimum overheads are introduced to the LP. P5. Only two extra bits for encoding four colors are needed for each node. Properties P1, P2, P3 and P4 are important to the performance of the system. But none of the previous systems satisfy all the four proper ties. (The system of Steele(1975) does not satisfy P1 and the system proposed by Di jk stra(1976) does not satisfy P2 and P3.) We have analyzed the performance of the parallel garbage collection system proposed in this paper. The results show that the parallel system is usually significantly more efficient in terms of storage and time than the sequential stack algorithm (see Appendix II). The comp lete performance results are reported in another paper (Kung and Song(1977)). In this paper we shall be mainly concerned with the correc tness of the system. We must be sure that a system is correct before studying its performance. The correctness proof In this paper Is not intended to be formal or completely rigorous. Given the complexity of the system, we feel that it is m...
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.