paperbot · PL 论文追踪

全部论文 待读 收藏

P2 Manticore: Hardware-Accelerated RTL Simulation with Static Bulk-Synchronous Parallelism

Manticore:基于静态整体同步并行的硬件加速 RTL 仿真
ASPLOS Proceedings of the 28th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 42023匹配度 1/10 · 该论文研究硬件加速的 RTL 仿真与静态整体同步并行,属于体系结构/并行仿真方向,与用户关注的 PL 与 AI 交叉、静态分析、浮点程序分析、贝叶斯程序分析及 LLM/Agent 均无直接关系。引用 14 inbox
Mahyar Emami, Sahand Kashani, Keisuke Kamahori, Mohammad Sepehr Pourghannad, Ritik Raj, James R. Larus
crossref: 2026-09-12

AI 速览

一句话:提出 Manticore,用硬件加速与静态整体同步并行加速周期精确 RTL 仿真,以缩小与硬件三个数量级的速度差距。

摩尔定律与 Dennard 缩放终结使专用计算机体系结构和加速器重新受到关注,而这类硬件的验证与测试高度依赖寄存器传输级(RTL)设计的周期精确仿真。当前最快的软件 RTL 仿真器吞吐率仅为 1–1000 kHz,比硬件实际运行慢三个数量级以上,成为设计迭代与设计空间探索的瓶颈。针对该问题,Manticore 采用硬件加速与静态整体同步并行(BSP)来实现 RTL 仿真,目标是提高仿真速度与设计者生产力。摘要以仿真 kHz 吞吐率为核心衡量指标,并与现有最快软件 RTL 仿真器及硬件速度进行对比,但未报告 Manticore 的具体实验数据。

问题
专用体系结构与加速器的验证测试依赖周期精确 RTL 仿真,但最快软件 RTL 仿真器仅达 1–1000 kHz,比硬件慢三个数量级以上,严重制约设计迭代与探索。
方法
Manticore 采用硬件加速与静态整体同步并行(BSP)来执行 RTL 仿真,以提升仿真吞吐率。
评估
摘要以仿真吞吐率(kHz)为衡量指标,将现有最快软件 RTL 仿真器与硬件速度进行对比;未给出 Manticore 的具体实验结果。
原文摘要(Abstract)

The demise of Moore's Law and Dennard Scaling has revived interest in specialized computer architectures and accelerators. Verification and testing of this hardware depend heavily upon cycle-accurate simulation of register-transfer-level (RTL) designs. The fastest software RTL simulators can simulate designs at 1--1000 kHz, i.e., more than three orders of magnitude slower than hardware. Improved simulators can increase designers' productivity by speeding design iterations and permitting more exhaustive exploration.

链接与引用

DOI 原文 · PDF(开放获取) ·

BibTeX
@inproceedings{paperbot10112,
  title = {Manticore: Hardware-Accelerated RTL Simulation with Static Bulk-Synchronous Parallelism},
  author = {Mahyar Emami and Sahand Kashani and Keisuke Kamahori and Mohammad Sepehr Pourghannad and Ritik Raj and James R. Larus},
  booktitle = {ASPLOS},
  year = {2023},
  doi = {10.1145/3623278.3624750}
}