arXiv:2506.06773cs.ARcs.LG2025-06中稿 · and presented at t…被引 1

针对难预测分支设计专用预测器,显著降低处理器误判率。

Taming Wild Branches: Overcoming Hard-to-Predict Branches using the Bullseye Predictor

  • 用专用子系统识别难预测分支并分派至特定感知机。
  • 实现平均每千条指令3.4045次误判,性能提升明显。
  • 适合关注处理器优化与分支预测的工程师与研究者。

分支预测是乱序处理器性能的关键。尽管CBP-2016冠军方案TAGE-SC-L结合了几何历史表、统计修正器和循环预测器,仍有超过一半的误判源于少数难预测(H2P)分支。这些分支在多样全局历史中出现,导致TAGE反复震荡并提前淘汰,使计数器无法成熟。先前工作表明扩大表格仅带来边际改善。本文在159 KB的TAGE-SC-L基础上,加入一个28 KB的专用于H2P分支的Bullseye预测器。它通过集合关联的H2P识别表(HIT)定位问题程序计数器(PC),并将它们导向两个分支特异的感知机:一个基于哈希局部历史,另一个基于折叠全局历史。一个短试验阶段在H2P缓存中追踪两者精度。只有当感知机持续准确且输出幅度超过动态阈值时,该分支才被纳入感知机,并抑制对TAGE的更新以减少污染。HIT、缓存与感知机与TAGE-SC-L并行运行,显著提升了对H2P尾部的预测精度。最终实现平均每千条指令3.4045次误判(MPKI)与145.09的周期加权每千条指令误判数(CycWpPKI)。

原文摘要 · Abstract (English)

Branch prediction is key to the performance of out-of-order processors. While the CBP-2016 winner TAGE-SC-L combines geometric-history tables, a statistical corrector, and a loop predictor, over half of its remaining mispredictions stem from a small set of hard-to-predict (H2P) branches. These branches occur under diverse global histories, causing repeated thrashing in TAGE and eviction before usefulness counters can mature. Prior work shows that simply enlarging the tables offers only marginal improvement. We augment a 159 KB TAGE-SC-L predictor with a 28 KB H2P-targeted subsystem called the Bullseye predictor. It identifies problematic PCs using a set-associative H2P Identification Table (HIT) and steers them to one of two branch-specific perceptrons, one indexed by hashed local history and the other by folded global history. A short trial phase tracks head-to-head accuracy in an H2P cache. A branch becomes perceptron-resident only if the perceptron's sustained accuracy and output magnitude exceed dynamic thresholds, after which TAGE updates for that PC are suppressed to reduce pollution. The HIT, cache, and perceptron operate fully in parallel with TAGE-SC-L, providing higher fidelity on the H2P tail. This achieves an average MPKI of 3.4045 and CycWpPKI of 145.09.

分支预测处理器优化硬件加速

Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品;中文卡片由大模型生成,请以原文为准。