POPL 2020 (series) / Student Research Competition /
Mechanization of Data Race Freedom guarantee proofs for Weakestmo memory model
It’s often very challenging for programmers to understand the semantics of concurrent code. The goal of our project is to verify the DRF properties for the Weakestmo memory model. For a programmer, it means that avoiding data races weaker than SC (or any data races) in his/her code is enough to rule out all the undesirable weak behaviors. Thus, one can abstract from the weak memory effects and reason as if it were sequentially consistent.