HN-F Snoop Filter 性能数学建模与分析 模型说明：本模型以单 Slice 覆盖率 $R_{\text{cover}}=C_{\text{SF,slice}}/C_{\text{L2}}$ 为自变量，所有性能量（溢出率、Miss 比例、延时、单核 IPC、相对衰减）均由此决定；核数 $N_{\text{core}}$ 仅作为整机总容量与 Aggregate IPC 的线性缩放因子出现。下文图表与结论均以单 Slice / 单核归一化量给出，对任意核数体系成立。 ...
核心结论：Snoop Filter（SF）的有效跟踪容量应配置为全系统 RN-F 缓存总和的 1.5× ~ 2×。这一结论并非工程经验值的外推，而是由组相联结构下的泊松分布尾部概率严格导出。SF 的容量直接决定了全系统缓存的有效利用率上限。 ...
一个懂 SystemVerilog、会拼位宽、能显示 12'hABC 的程序员计算器。开源了。 做芯片这行，日常工作里少不了跟进制、位宽、bit mask 打交道。读 AXI 总线地址、手算寄存器域段、验证 CRC 结果——随时都要切进制、移位、拼位宽。 ...
PCIE AMBA Integration Guide 1. Introduction 1.1. Introduction This document is intended to provide guidance on PCIe interface integration into an AMBA based System-On-Chip (SoC). It is assumed that the PCIe interface is connected to the rest of the SoC through an AXI or ACE protocol- based interconnect. The reader is assumed to be conversant with the PCIe, AMBA AXI and AMBA ACE protocols and Arm architecture. 本文档旨在为 PCIe 接口集成到基于 AMBA 的片上系统 (SoC) 提供指导。假设 PCIe 接口通过基于 AXI 或 ACE 协议的互连连接到 SoC 的其余部分。读者应熟悉 PCIe、AMBA AXI 和 AMBA ACE 协议以及 Arm 架构。 ...
AMBA CHI Chip-to-Chip Architecture Specication AMBA CHI Chip-to-Chip (C2C) Architecture Specification 1. Introduction This specification is a Chip-to-Chip (C2C) extension of the AMBA CHI Architecture Specification. 1.1. Overview The C2C extension enables building a system with multiple CPUs, accelerators, or other devices. CHI C2C focuses on the Packetization of CHI messages, making them suitable to be transported over a chip-to-chip link. The Packetization format is optimized for link utilization and latency, while avoiding complex packing and unpacking schemes. ...