by Timothy H. Heil, Zak Smith, J. E. Smith
http://web.demigod.org/~zak/research/heil-micro-32.ps
Add To MetaCart
Abstract:
Branch predictors typically use combinations of branch PC bits and branch histories to make predictions. Recent improvements in branch predictors have come from reducing the effect of interference, i.e. multiple branches mapping to the same table entries. In contrast, the branch difference predictor (BDP) uses data values as additional information to improve the accuracy of conditional branch predictors. The BDP maintains a history of differences between branch source register operands, and feeds these into the prediction process. An important component of the BDP is a rare event predictor (REP) which reduces learning time and table interference. An REP is a cache-like structure designed to store patterns whose predictions differ from the norm. Initially, ideal interference-free predictors are evaluated to determine how data values improve correlation. Next, execution driven simulations of complete designs realize this potential. The BDP reduces the misprediction rate of five SPEC95 integer benchmarks by up to 33 % compared to gshare and by up to 15 % compared to Bi-Mode predictors. 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
|
|
246
|
Exceeding the Dataflow Limit via Value Prediction
– Lipasti, Shen
- 1996
|
|
227
|
A Comparison of Dynamic Branch Predictors that Use Two Levels of Branch History
– Yeh, Patt
- 1993
|
|
221
|
The predictability of data values
– Sazeides, Smith
- 1997
|
|
172
|
Highly Accurate Data Value Prediction using Hybrid Predictors
– Wang, Franklin
- 1997
|
|
100
|
Implementing the ppm data compression scheme,” in
– Moffat
- 1990
|
|
95
|
The YAGS branch prediction scheme
– Eden, Mudge
- 1998
|
|
86
|
The Bi-Mode Branch Predictor
– Lee, Chen, et al.
- 1997
|
|
83
|
Streamlining Interoperation Memory Communication via Data Dependence Prediction
– Moshovos, Sohi
- 1997
|
|
72
|
Dynamic Path-Based Branch Correlation
– Nair
- 1995
|
|
65
|
Analysis of branch prediction via data compression
– Chen, Coffey, et al.
- 1996
|
|
63
|
Dynamic History-Length Fitting: A third level of adaptivity for branch prediction
– Juan, Sanjeevan, et al.
- 1998
|
|
59
|
Correlation and Aliasing in Dynamic Branch Predictors
– Sechrest, Lee, et al.
- 1996
|
|
55
|
Alternative implementations of hybrid branch predictors
– Chang, Hao, et al.
- 1995
|
|
41
|
Efficacy and Performance Impact of Value Prediction
– Rychlik, Faistl, et al.
- 1998
|
|
39
|
Improving CC-NUMA Performance Using Instruction-Based Prediction
– Kaxiras, Goodman
- 1999
|
|
38
|
Dataflow Analysis of Branch Mispredictions and Its Application to Early Resolution of Branch Outcomes
– Farcy, Temam, et al.
- 1998
|
|
35
|
Variable length path branch prediction
– Stark, Evers, et al.
- 1998
|
|
25
|
Architectural support for compiler-synthesized dynamic branch prediction strategies: Rationale and initial results
– August, Connors, et al.
- 1997
|
|
23
|
Control-flow speculation through value prediction for superscalar processors
– González, González
- 1999
|
|
17
|
Predicting Indirect Branches via Data Compression
– Kalamatianos, Kaeli
- 1998
|
|
13
|
Trading Conflict and Capacity Aliasing
– Michaud, Seznec, et al.
- 1997
|
|
5
|
Difference Prediction and the Rare Event Predictor
– Heil, “Branch
- 1999
|
|
1
|
Intel Discloses New IA-64 Features," Microprocessor Report
– Gwenmap
- 1999
|
