MemorymodelformultithreadedC++.PDFVIP

Document Number: WG21/N1680=J16/04-0120 Date: 10 September 2004 Reply to: Andrei Alexandrescu andrei@ 2216 NE 46th St. Apt. A Seattle WA 98105 USA Memory model for multithreaded C++ Andrei Alexandrescu Hans Boehm Kevlin Henney Doug Lea Bill Pugh Abstract The C++ Standard de?nes single-threaded program execution. Fun- damentally, multithreaded execution requires a much more re?ned mem- ory and execution model. C++ threading libraries are in the awkward sit- uation of specifying (implicitly or explicitly) an extended memory model for C++ in order to specify program execution. We propose integrating a memory model suitable for multithreaded execution in the C++ Stan- dard. On top of that model, we propose a standard threading library. 1 Introduction Many of today’s applications make use of multithreaded execution. We expect such use to grow as the increased use of hardware multithreading (a.k.a. “hy- perthreading”) and multi-core processors will force or entice more and more applications to become multithreaded. C++ is commonly used as part of mul- tithreaded applications, sometimes with either direct calls into an OS-provided threading library (e.g. POSIX threads (pthreads) [3] or Win32 threads) or with the aid of an intervening layer that provides a platform-neutral interface (e.g. Boost Threads). However, concurrent programming is inherently diverse, and includes high-performance parallel programming, asynchronous task processing, message-based and event-based systems, non-blocking, lock-free and optimistic data structures, transactional approaches, and so on. Unfortunately, progra