Elastic History Buffer: A Low-Cost Method to Improve Branch Prediction Accuracy
نویسندگان
چکیده
Two-level dynamic branch predictors try to predict the outcomes of conditional branches using both a table of state counters associated with specific branch instructions and a buffer of recent branch outcomes to correlate the counters with specific branch histories. However, there is always a question of how much correlation to use, and some programs benefit from higher levels of correlation than others. This paper presents the Elastic History Buffer (EHB), a low-cost yet effective scheme that can exploit the property that each branch instruction may have a different degree of correlation with other branches, while keeping the simple structure of a single global branch history. We have simulated the EHB on SPECint92 for two architectures. On average, the EHB has 25% fewer mispredictions than fixed-correlation schemes and 10% fewer than frequencybased branch classification schemes. With limited hardware (1KB), the EHB is close to the optimum measured by repeating the experiments on an “oracle” two-level predictor.
منابع مشابه
An Integrated Placement and Synthesis Approach for Timing Closure of PowerPC Microprocessors
Power Compiler: A Gate-Level Power Optimization and Synthesis System p. 74 Elastic History Buffer: A Low-Cost Method to Improve Branch Prediction Accuracy p. 82 Design Optimization for High-Speed Per-address Two-level Branch Predictors p. 88 PA-8000: A Case Study of Static and Dynamic Branch Prediction p. 97 Discrete Drive Selection for Continuous Sizing p. 110 Continuous Retiming: Algorithms a...
متن کاملElastic History Buuer: a Low-cost Method to Improve Branch Prediction Accuracy I Contents 1 Introduction 1 2 Branch Prediction: Previous Work and Problem Statement 2
As computer architects try to squeeze more performance out of superscalar processors by increasing issue widths and pipeline depths, the penalties for incorrectly predicting conditional branches will continue to rise. Current state-of-the-art processors use a table of state counters associated with speci c branches, coupled with a bu er recording recent branch outcomes to correlate the counters...
متن کاملThe Cascaded Predictor: Economic and Adaptive Branch Target Prediction
Two-level predictors improve branch prediction accuracy by allowing predictor tables to hold multiple predictions per branch. Unfortunately, the accuracy of such predictors is impaired by two detrimental effects. Capacity misses increase since each branch may occupies entries proportional to the number of different path histories leading up to the branch. The working set of a given program ther...
متن کاملBehavior-Based Branch Prediction by Dynamically Clustering Branch Instructions
Conditional branches frequently exhibit similar behavior (bias, time-varying behavior, ...), a property that can be used to improve branch prediction accuracy. Branch clustering constructs groups or clusters of branches with similar behavior and applies different branch prediction techniques to each branch cluster. We revisit the topic of branch clustering with the aim of generalizing branch cl...
متن کاملDynamic per-branch history length adjustment to improve branch prediction accuracy
Branch misprediction limits processor performance signiWcantly, as the pipeline deepens and the instruction issued per cycle increases. Since the introduction of the two-level adaptive branch predictor, branch history has been a major input vector in branch prediction, together with the address of a branch instruction. Until now, the length of branch history has been statically Wxed for all bra...
متن کامل