The E ects of Mispredicted - Path Execution on BranchPrediction

Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works, must be obtained from the IEEE. Abstract Branch prediction accuracies determined using trace-driven simulation do not include the eeects of executing branches along a mispredicted path. However, branches along a mispredicted path will pollute the branch prediction structures if no recovery mechanisms are provided. Without recovery mechanisms, prediction rates will suuer. In this paper, we determine the appropriateness of recovery mechanisms for the four structures of the Two-Level Adaptive Branch Predic-tor: the Branch Target Buuer (BTB), the Branch History Register (BHR), the Pattern History Tables (PHTs), and the Return Address Stack (RAS). We then propose cost-eeective recovery mechanisms for these branch prediction structures. For ve benchmarks from the SPECint92 suite, we show that performance is not aaected if recovery mechanisms are not provided for the BTB and the PHTs. On the other hand, without any recovery mechanisms for the BHR and RAS, performance drops by an average of 29%.