Enveloping implicit assumptions of intrusive data structures within ownership type system

Abstract

Intrusive data structures (IDSes) are heavily used in system programming, where achieving high performance is one of the most important design goals. Yet, they are not supported in today’s ownership type system that offer memory-safety without garbage collection. Instead, IDSes force programmers to choose either unsafety or runtime overhead. This limitation stems from the implicit assumptions pertaining to the memory layouts and access patterns created by IDSes. In this paper, we propose a new technique, referred to as ownership pooling, which defines ownership for IDSes. Ownership pooling consists of three new types, FieldOf, OwnershipPool, and Shared, and their conversion rules. We implemented the proposed types within Rust’s type system and compared its performance capabilities against the existing memory-safe implementations and the C++ implementation without memory safety as baseline. The performance of our implementation shows far better performance than that of the existing memory-safe ones and comparable to that of C++ implementation without memory safety.