An All-Software Thread-Level Data Dependence Speculation System for Multiprocessors
作者： Per Stenström , Peter Rundberg
DOI:
关键词:
摘要: We present a software approach to design thread-level data dependence speculation system targeting multiprocessors. Highly-tuned checking codes are associated with loads and stores whose addresses cannot be disambiguated by parallel compilers that can potentially cause violations at run-time. Besides resolving many name true dependencies through dynamic renaming forwarding, respectively, our method supports commit operations. Performance results collected on an architectural simulator validated commercial multi-processor show the overhead reduced less than ten instructions per speculative memory operation. Moreover, we demonstrate ten-fold speedup is possible some of difficult-to-parallelize loops in Perfect Club benchmark suite 16-way multiprocessor.
chalmers.se
jilp.org 参考文章(22)
Stanford University. Computer Systems Laboratory, Software and Hardware for Exploiting Speculative Parallelism with a Multiprocessor Stanford University. ,(1997)
John L. Hennessy, David A. Patterson, Computer Architecture: A Quantitative Approach ,(1989)
Bengt Werner, Fredrik Larsson, Peter S. Magnusson, Fredrik Lundholm, Magnus Karlsson, Andreas Moestedt, Per Stenström, Fredrik Dahlgren, Jim Nilsson, Håkan Grahn, SimICS/sun4m: a virtual workstation usenix annual technical conference. pp. 10- 10 ,(1998)
Jenn-Yuan Tsai, Pen-Chung Yew, The superthreaded architecture: thread pipelining with run-time data dependence checking and control speculation international conference on parallel architectures and compilation techniques. pp. 35- 46 ,(1996) , 10.1109/PACT.1996.552553
Ye Zhang, L. Rauchwerger, J. Torrellas, Hardware for speculative run-time parallelization in distributed shared-memory multiprocessors high-performance computer architecture. pp. 162- 173 ,(1998) , 10.1109/HPCA.1998.650556
J.G. Steffan, T.C. Mowry, The potential for using thread-level data speculation to facilitate automatic parallelization high-performance computer architecture. pp. 2- 13 ,(1998) , 10.1109/HPCA.1998.650541
Iffat H. Kazi, David J. Lilja, Coarse-grained speculative execution in shared-memory multiprocessors international conference on supercomputing. pp. 93- 100 ,(1998) , 10.1145/277830.277853
Blu William, Ramon Doallo, Rudolf Eigenmann, John Grout, Jay Hoeflinger, Thomas Lawrence, Jaejin Lee, David Padua, Yunheung Paek, Bill Pottenger, Lawrence Rauchwerger, Peng Tu, Parallel programming with Polaris IEEE Computer. ,vol. 29, pp. 78- 82 ,(1996) , 10.1109/2.546612
Pedro Marcuello, Antonio González, Clustered speculative multithreaded processors international conference on supercomputing. pp. 365- 372 ,(1999) , 10.1145/305138.305214
Venkata Krishnan, Josep Torrellas, Hardware and software support for speculative execution of sequential binaries on a chip-multiprocessor international conference on supercomputing. pp. 85- 92 ,(1998) , 10.1145/277830.277852