AmpereOne CPU 核心的性能验证
Performance Verification of the AmpereOne CPU Core
AI总结：
该研究针对工艺缩放放缓背景下的性能验证需求，提出应用于四代 AmpereOne CPU 核心的工业级分层性能验证方法论，可避免硅后错误并保障处理器性能目标。
AI中文摘要：
随着工艺技术缩放速度放缓，微架构创新已成为性能提升的主要驱动力，这使得硅前性能验证（PV）比以往任何时候都更为关键。本文介绍了应用于四代 AmpereOne 定制 CPU 核心的工业级 PV 方法论，核心是 RTL 设计与追踪驱动性能模型的周期级精确关联。该方法论整合了数据驱动的工作负载精选、高频每日回归系统以及用于分析的统一事件流框架。我们通过分支预测单元和 L2 预取器的案例研究展示了该方法论，强调了一种分层策略：首先隔离单个单元进行聚焦关联，再进行全核心验证。结果表明，这种严谨的迭代过程对于避免代价高昂的硅后错误、确保复杂处理器达到性能目标不可或缺。最后，我们展望了微处理器行业 PV 的未来。
英文摘要：
As process technology scaling slows, microarchitectural innovation has become the primary driver of performance gains, making pre-silicon Performance Verification (PV) more critical than ever. This paper presents the industrial-scale PV methodology applied across four generations of the AmpereOne custom CPU core, centered on the cycle-accurate correlation of the RTL design against a trace-driven performance model. The methodology integrates data-driven workload curation, a high-frequency daily regression system, and a unified event-stream framework for analysis. We demonstrate this methodology through case studies of the Branch Prediction Unit and L2 Prefetcher, highlighting a hierarchical strategy that first isolates individual units for focused correlation before proceeding to full-core verification. The results demonstrate that this disciplined, iterative process is indispensable for avoiding costly post-silicon bugs and ensuring complex processors meet their performance targets. We end with a look towards the future of PV in the microprocessor industry.
