A Separation Logic for a Promising Semantics

Abstract

We present SLR, the first expressive program logic for reasoning about concurrent programs under a weak memory model addressing the out-of-thin-air problem. Our logic includes the standard features from existing logics, such as RSL and GPS, that were previously known to be sound only under stronger memory models: (1) separation, (2) per-location invariants, and (3) ownership transfer via release-acquire synchronisation-as well as novel features for reasoning about (4) the absence of out-of-thin-air behaviours and (5) coherence. The logic is proved sound over the recent "promising" memory model of Kang et al., using a substantially different argument to soundness proofs of logics for simpler memory models.

On library correctness under weak memory consistency: specifying and verifying concurrent libraries under declarative consistency models

“…On the program logic side, there is a number of logics for reasoning about different fragments of the (R)C11 memory model Vafeiadis 2016, 2017;Kaiser et al 2017;Svendsen et al 2018;Tassarotti et al 2015;Turon et al 2014;Vafeiadis and Narayan 2013], the x86-TSO model [Ridge 2010;Sieczkowski et al 2015], as well as a logic that is parametrised over the memory model [Alglave and Cousot 2017]. While most of these logics provide facilities 68:29 for writing abstract specifications of concurrent libraries, their specification language is not rich enough to express the functional correctness and atomicity specification of a concurrent stack or a queue.…”

Grounding thin-air reads with event structures

“…defined the promising semantics, which comes with a number of results, which we are able to exploit for our weakest model. It has also led to some follow up work, which proved correctness of compilation to ARM [Podkopaev et al 2017[Podkopaev et al , 2019 and the correctness of a program logic over it [Svendsen et al 2018]. Nevertheless, as discussed, the promising semantics also has a number of shortcomings, which are difficult to resolve because of the model's complexity and brittleness.…”

“…Their soundness proofs, however, rely heavily on the acyclicity of po ∪ rf and it is unclear whether they can be adapted to the weaker setting of weakestmo. Among these logics, RSL [Vafeiadis and Narayan 2013] should be sound under our model, because Svendsen et al [2018] proved the soundness of a variant of it over the promising semantics. For the more advanced logics, however, the counterexample of Doko and Vafeiadis [2017] shows that certain adaptations will be needed.…”

RustBelt meets relaxed memory

“…Kang et al 's promising semantics [Kang et al 2017] is a proposal for fixing C11's out-of-thin-air problem without prohibiting load-store reordering on relaxed accesses (as RC11 and ORC11 do). Svendsen et al [2018] introduce the first program logic for the promising semantics. Their logic is based on RSL [Vafeaidis and Narayan 2013] and supports relaxed accesses but not fences.…”