5. Concluding Remarks 4.3 Breakdown of Misprediction for the Gshare and Bi-mode Schemes 4.4 Go Benchmark Table 4: Numbers of Changes between Different Bias Classes for the History-indexed and Bi-mode Schemes

@inproceedings{Mudge19925CR,
  title={5. Concluding Remarks 4.3 Breakdown of Misprediction for the Gshare and Bi-mode Schemes 4.4 Go Benchmark Table 4: Numbers of Changes between Different Bias Classes for the History-indexed and Bi-mode Schemes},
  author={Trevor N. Mudge},
  year={1992}
}
schemes and configurations the misprediction for the WB class dominates—destructive aliasing is not the major concern. There is not much room for the bi-mode scheme to improve because it is targeted at eliminating harmful aliasing rather than improving prediction for the weakly biased sub-streams. As observed in the previous subsection, the dynamic frequency of the weakly biased class is mainly determined by the number of global history bits used. From Figure 8, we see that the error of the WB… CONTINUE READING