MetaCartSign in to MyCiteSeer

Include Citations | Advanced Search | Help

Include Citations | Advanced Search | Help

  Novel Techniques to Improve Branch Prediction Accuracy for Embedded Processors in the Presence of Context Switches

Download:
Download as a PDF
by Sudeep Pasricha, Alex Veidenbaum
http://www.cecs.uci.edu/technical_report/TR03-24.pdf
Add To MetaCart

Abstract:

Embedded processors like Intel’s XScale use dynamic branch prediction to improve performance. Due to the presence of context switches, the accuracy of these predictors is reduced because they end up storing prediction histories for several processes. This paper shows that the loss in accuracy can be significant and depends on predictor type and size. Several new schemes are proposed to save and restore the predictor state on context switches in order to improve prediction accuracy. The schemes differ in the amount of information they save and vary in their accuracy improvement. It is shown that even for a small 1K entry hybrid predictor, 1- 6 % improvement in prediction rate can be achieved (for an average context switch interval of 100K instructions) for several embedded applications while saving and restoring a minimal amount of state information (less than 128 bits) on a context switch. 1.

Citations

1253 The Simplescalar toolset, version 2.0 – Burger, Austin - 1997
521 Combining branch predictors – McFarling - 1993
365 A Study of Branch Prediction Strategies – Smith - 1981
231 Alternative implementations of twolevel adaptive branch prediction – Yeh, Patt - 1992
216 Branch Prediction Strategies and Branch Target Buffer Design – Lee, Smith - 1984
215 Mibench: A free, commercially representative embedded benchmark suite – Guthaus, Ringenberg, et al.
172 Improving the accuracy of dynamic branch prediction using branch correlation – Pa, So, et al. - 1992
95 The YAGS branch prediction scheme – Eden, Mudge - 1998
89 Using Hybrid Branch Predictors to Improve Branch Prediction Accuracy in the Presence of Context Switches – Evers, Chang, et al.
86 The Bi-Mode Branch Predictor – Lee, Chen, et al. - 1997
85 Branch classification: a new mechanism for improving branch predictor performance – Chang, Hao, et al. - 1994
76 Trading conflict and capacity aliasing in conditional branch predictors – Seznec - 1997
72 Dynamic Path-Based Branch Correlation – Nair - 1995
63 Dynamic History-Length Fitting: A third level of adaptivity for branch prediction – Juan, Sanjeevan, et al. - 1998
55 Improving Branch Prediction Accuracy by Reducing Pattern History Table Interference – Chang, Evers, et al. - 1996
55 Branch target buffer design and optimization – Perleberg, Smith - 1989
8 An Analysis of Dynamic Branch – Gloy, Young, et al. - 1996
4 Saving and restoring implementation contexts with co-designed virtual machines – Dhodapkar, Smith - 2005
3 Skadron “The Effects of Context Switching on Branch Predictor Performance – Co, K - 2001
1 et al., "A taxonomy of branch mispredictions, and alloyed prediction as a robust solution to wrong-history mispredictions – Skadron