Linearizable Implementations Suffice for Termination of Randomized Concurrent Programs

Strong adversaries obtain additional power when a linearizable object is substituted instead of an atomic object in a concurrent program. This paper suggests a novel approach to blunting this additional power, without relying on strongly linearizable implementations. Instead, a simple modification of some existing linearizable implementations is proposed with the property that if a concurrent program has non-zero termination probability when used with atomic objects, then it also has non-zero termination probability when it is used with the modified linearizable implementations. Our results apply to the ABD implementation of a shared register in asynchronous message-passing systems and also to AAD+ linearizable snapshots in asynchronous shared-memory systems.