Incorporation of OpenMP memory consistency into conventional dataflow analysis

Authors:
Ayon Basumallik;Rudolf Eigenmann
Affiliations:
School of Electrical and Computer Engineering, Purdue University, West Lafayette, IN;School of Electrical and Computer Engineering, Purdue University, West Lafayette, IN
Venue:
IWOMP'08 Proceedings of the 4th international conference on OpenMP in a new era of parallelism
Year:
2008

Citing 12
Cited 2

Efficient and correct execution of parallel programs that share memory

ACM Transactions on Programming Languages and Systems (TOPLAS)
Analyses and optimizations for shared address space programs

Journal of Parallel and Distributed Computing - Special issue on compilation techniques for distributed memory systems
Hiding Relaxed Memory Consistency with a Compiler

IEEE Transactions on Computers - Special issue on the parallel architecture and compilation techniques conference
A Unified Formalization of Four Shared-Memory Models

IEEE Transactions on Parallel and Distributed Systems
Automatic fence insertion for shared memory multiprocessing

ICS '03 Proceedings of the 17th annual international conference on Supercomputing
A compiler for multiple memory models: Research Articles

Concurrency and Computation: Practice & Experience - Compilers for Parallel Computers
Towards automatic translation of OpenMP to MPI

Proceedings of the 19th annual international conference on Supercomputing
Parallel Data Flow Analysis for OpenMP Programs

IWOMP '07 Proceedings of the 3rd international workshop on OpenMP: A Practical Programming Model for the Multi-Core Era
Static nonconcurrency analysis of OpenMP programs

IWOMP'05/IWOMP'06 Proceedings of the 2005 and 2006 international conference on OpenMP shared memory parallel programming
The OpenMP memory model

IWOMP'05/IWOMP'06 Proceedings of the 2005 and 2006 international conference on OpenMP shared memory parallel programming
Formal specification of the OpenMP memory model

IWOMP'05/IWOMP'06 Proceedings of the 2005 and 2006 international conference on OpenMP shared memory parallel programming
Automatic scoping of variables in parallel regions of an OpenMP program

WOMPAT'04 Proceedings of the 5th international conference on OpenMP Applications and Tools: shared Memory Parallel Programming with OpenMP

There is nothing wrong with out-of-thin-air: compiler optimization and memory models

Proceedings of the 2011 ACM SIGPLAN Workshop on Memory Systems Performance and Correctness
ompVerify: polyhedral analysis for the OpenMP programmer

IWOMP'11 Proceedings of the 7th international conference on OpenMP in the Petascale era

Quantified Score

Hi-index	0.00

Visualization

Abstract

Current OpenMP compilers are often limited in their analysisand optimization of OpenMP programs by the challenge of incorporatingOpenMP memory consistency semantics into conventional dataflow algorithms. An important reason for this is that data flow analysiswithin current compilers traverse the program's control-flow graph(CFG) and the CFG does not accurately model the memory consistencyspecifications of OpenMP. In this paper, we present techniquesto incorporate memory consistency semantics into conventional dataflowanalysis by transforming the program's CFG into an OpenMP Producer-Consumer Flow Graph (PCFG), where a path exists from writes to readsof shared data if and only if a dependence is implied by the OpenMPmemory consistency model. We present algorithms for these transformations,prove the correctness of these algorithms and discuss a case wherethis transformation is used.