A Comparison Of Dynamic Branch Predictors That Use Two Levels Of Branch History

Recent attention to speculative ezecution as a mechanism for increasing performance of single instruction streams has demanded substantially better branch prediction than what has been previously available. We [l, 21 and Pan, So, and Rahmeh [4] have both proposed variations of the same aggressive dynamic branch predictor for handling those needs. We call the basic model Two-Level Adaptive Branch Prediction; Pan, So, and Rahmeh call it Correlation Branch Prediction. In this paper, we adopt the terminology of [2] and show that there are really nine variations of the same basic model. W e compare the nine variations with respect to the amount of history information kept. W e study the effects of different branch history lengths and pattern history table configurations. Finally, we evaluate the cost effectiveness of the nine variations.