Relaxed memory models: an operational approach

Gérard Boudol 1, 2 Gustavo Petri 1, 2
1 MIMOSA - Migration and mobility : semantics and applications
CRISAM - Inria Sophia Antipolis - Méditerranée , Université de Provence - Aix-Marseille 1, MINES ParisTech - École nationale supérieure des mines de Paris
2 INDES - Secure Diffuse Programming
CRISAM - Inria Sophia Antipolis - Méditerranée
Abstract : Memory models define an interface between programs written in some language and their implementation, determining which behaviour the memory (and thus a program) is allowed to have in a given model. A minimal guarantee memory models should provide to the programmer is that well-synchronized, that is, data-race free code has a standard semantics. Traditionally, memory models are defined axiomatically, setting constraints on the order in which memory operations are allowed to occur, and the programming language semantics is implicit as determining some of these constraints. In this work we propose a new approach to formalizing a memory model in which the model itself is part of a weak operational semantics for a (possibly concurrent) programming language. We formalize in this way a model that allows write operations to the store to be buffered. This enables us to derive the ordering constraints from the weak semantics of programs, and to prove, at the programming language level, that the weak semantics implements the usual interleaving semantics for data-race free programs, hence in particular that it implements the usual semantics for sequential code.
Complete list of metadatas

Cited literature [32 references]  Display  Hide  Download

https://hal.inria.fr/inria-00420352
Contributor : Gustavo Petri <>
Submitted on : Monday, September 28, 2009 - 4:54:12 PM
Last modification on : Tuesday, November 13, 2018 - 1:15:42 AM
Long-term archiving on : Wednesday, June 16, 2010 - 12:11:27 AM

File

gg-popl.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : inria-00420352, version 1

Citation

Gérard Boudol, Gustavo Petri. Relaxed memory models: an operational approach. POPL'09, Jan 2009, Savannah, GA,, United States. ⟨inria-00420352⟩

Share

Metrics

Record views

190

Files downloads

266