Safety of Live Transactions in Transactional Memory: TMS is Necessary and Sufficient

The main challenge of stating the correctness of transactional memory (TM) systems is the need to provide guarantees on the system state observed by live transactions, i.e., those that have not yet committed or aborted. A TM correctness condition should not be too restrictive, to allow flexibility in implementation, yet strong enough to disallow undesirable TM behavior, which can lead to run-time errors in live transactions. The latter feature is formalized by observational refinement between TM implementations, stating that properties of a program using a concrete TM implementation can be established by analyzing its behavior with an abstract TM, serving as a specification of the concrete one. We show that a variant of transactional memory specification (TMS) is equivalent to observational refinement for the common programming model in which local variables are rolled back upon a transaction abort and, hence, is the weakest consistency condition for this case. This is challenging due to the nontrivial formulation of TMS, which allows different aborted and live transactions to have different views on the system state. Our proof reveals some natural, but subtle, assumptions on the TM required for the equivalence result. This is a submission to the regular track. It can be considered for the best student paper award: Sandeep Hans is a full-time student. ∗Technion Israel Institute of Technology, Israel, {hagit,sandeep}@cs.technion.ac.il †IMDEA Software Institute, Spain, [email protected] ‡Tel Aviv University, Israel, [email protected]