article · 1994
The effect of speculatively updating branch history on branch prediction accuracy, revisited
Eric Hao, Po-Yung Chang, Yale N. Patt
Cited by 66Open access
Abstract
Recent research has suggested that the branch history register need not contain the outcomes of the most recent branches in order for the Two-Level Adaptive Branch Predictor to work well. From this result, it is tempting to conclude that the branch history register need not be speculatively updated. This paper revisits this work and explains when the most recent branch outcomes can be omitted without significantly affecting performance. It also explains why this result does not imply that speculative update is not important. This paper shows that because the number of unresolved branches present in the machine varies during program execution, branch predictors without speculative update perform significantly worse than branch predictors with speculative update.
Cite this paper
Hao, E., Chang, P.-Y., & Patt, Y. N. (1994). The effect of speculatively updating branch history on branch prediction accuracy, revisited. https://doi.org/10.1145/192724.192756
Hao, Eric, et al. “The effect of speculatively updating branch history on branch prediction accuracy, revisited.” 1994, pp. 228–32, https://doi.org/10.1145/192724.192756.
Hao, Eric, Po-Yung Chang, and Yale N. Patt. 1994. “The effect of speculatively updating branch history on branch prediction accuracy, revisited.” https://doi.org/10.1145/192724.192756.
Hao, E., Chang, P.-Y. and Patt, Y.N. (1994) ‘The effect of speculatively updating branch history on branch prediction accuracy, revisited’. Available at: https://doi.org/10.1145/192724.192756.
@article{hao1994effect,
title = {{The effect of speculatively updating branch history on branch prediction accuracy, revisited}},
author = {Hao, Eric and Chang, Po-Yung and Patt, Yale N.},
pages = {228--232},
year = {1994},
doi = {10.1145/192724.192756}
}Made from the index record. Check names, capitalisation and page numbers against the paper. Need another style? AI for Research has over 10,000.
Related papers
Broadcast Mechanism for improving Conditional Branch Prediction in Speculative Multithreaded Processors
Renjith Retnamma · PDXScholar (Portland State University) · 2000
Cited by 0Open access
The Impact of Speculative Execution on SMT Processors
Dongsoo Kang, Chen Liu, Jean‐Luc Gaudiot · International Journal of Parallel Programming · 2007
Cited by 6
An Out-of-order Superscalar Processor With Speculative Execution And Fast, Precise Interrupts
H. Dwyer, H. C. Torng · IEEE/ACM International Symposium on Microarchitecture (MICRO) · 2005
Cited by 9
Modeling Control Speculation for Timing Analysis
Xianfeng Li, Tulika Mitra, Abhik Roychoudhury · Real-Time Systems · 2004
Cited by 37
Cost effective speculation with the omnipredictor
Arthur Pérais, André Seznec · International Conference on Parallel Architectures and Compilation Techniques (PACT) · 2018
Cited by 10Open access
The effect of speculative updating branch history on branch prediction accuracy, revisited
Eric Hao, Po-Yung Chang, Yale N. Patt · IEEE/ACM International Symposium on Microarchitecture (MICRO) · 1994
Cited by 62Open access