arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2607.01430cs.AR
收藏

物理感知的抢占式虚拟通道用于无死锁AXI网络片

Physically-Aware Preemptive Virtual Channels for Deadlock-Free AXI Networks-on-Chip

ETH Zürich(苏黎世联邦理工学院);Università di Bologna(博洛尼亚大学)

Lorenzo Leone, Luca Colagrande, Luca Benini

AI总结:

针对AXI4协议中读写流量依赖导致的死锁问题,提出物理感知的抢占式虚拟通道架构,相比多平面设计节省76%链路资源,仅增加3%路由器面积开销。

摘要:

随着众核片上系统(SoC)持续扩展,网络片(NoC)必须维持日益增长的内存带宽,同时保证无死锁。在AXI4系统中,即使路由算法本身无死锁,读写流量之间的协议级依赖也可能在网络端点产生循环等待。解耦这些流量类别可以避免此类依赖,但暴露了一个关键的实现权衡:多平面NoC复制宽物理链路并增加路由压力,而传统的虚拟通道(VC)路由器则增加了大量的控制复杂度、面积和时序开销。本文重新审视了现代宽链路NoC中的这一权衡。我们评估了四种无死锁的AXI4流量类别分离方案:一个多平面基线和三种轻量级基于VC的设计。在这些设计中,我们提出了抢占式VC,一种物理感知架构,与多平面设计相比,可节省高达76%的链路资源,同时频率相当,路由器面积开销仅增加3%。

英文摘要:

As many-core Systems-on-Chip (SoCs) continue to scale, Networks-on-Chip (NoCs) must sustain increasingly high memory bandwidth while preserving deadlock freedom. In AXI4 systems, protocol-level dependencies between read and write traffic can create circular waits at the network endpoints, even when the routing algorithm itself is deadlock-free. Decoupling these traffic classes avoids such dependencies, but exposes a key implementation trade-off: multiplane NoCs duplicate wide physical links and increase routing pressure, whereas conventional Virtual Channel (VC) routers add substantial control complexity, area, and timing overhead. This work revisits this trade-off for modern wide-link NoCs. We evaluate four deadlock-free AXI4 traffic-class separation schemes: a multiplane baseline and three lightweight VC-based designs. Among these designs, we propose Preemptive VCs, a physically-aware architecture that can save up to 76% of link resources with comparable frequency and only 3% router area overhead relative to the multiplane design.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑