Design and verification of bus bridge between TileLink and AXI4 based on RISC-V processor
-
摘要:
RISC-V是近年提出的一种开源精简指令集架构,TileLink总线是专为RISC-V处理器设计的片上总线.为使RISC-V处理器灵活适配更多已有的AXI4 IP资源,提出一种高效率TileLink与AXI4总线桥设计方案,其中由一系列功能子模块匹配总线间数据传输方式的差异,以流水线传输形式实现数据跨协议的传输,增加总线桥的数据吞吐量.在实现总线桥不同通道间的转换时,采用不同的仲裁策略,在AXI4总线的响应转换过程中,采用固定优先级仲裁,优先转换数据响应,保证系统整体运行效率;在AXI4总线的写数据和读数据事务转换过程中,采用轮询仲裁,保证写数据和读数据的公平性,均衡分配目标通道带宽,提高总线带宽利用率和系统传输效率.从模块级用TileLink随机测试激励对总线桥进行仿真验证,并通过在RISC-V处理器上挂载AXI4接口PCI Express根复合体,从FPGA系统级进行验证,结果表明,设计的总线桥能够正确转换协议,并且能较大提高系统带宽利用率.总线桥在SMIC 55 nm CMOS工艺下进行了ASIC实现,工作频率达714 MHz,版图面积405×405
μ m2.-
关键词:
- RISC-V /
- 总线桥 /
- TileLink总线 /
- AXI4总线 /
- 流水线传输
Abstract:RISC-V is an open reduced instruction set architecture in recent years, and TileLink is a chip-scale interconnect standard designed for RISC-V. In order to use the existing AXI4 IP (Intellectual Property) resources flexibly in RISC-V processors, an efficient bus bridge design scheme between TileLink and AXI4 is proposed. A series of sub-modules match the transaction differences between Tilelink and AXI4, and complete the data transmission across protocols in the form of pipeline transmission, increasing the data throughput of the bus bridge. Different arbitration strategies are used to realize the conversion between different channels of the bus bridge. In the process of AXI4 bus response conversion, fixed-priority arbitration is used to preferentially convert read response, which improves the operating efficiency of the system. In the process of AXI4 bus write and read transactions conversion, round-robin arbitration is used to ensure the fairness of write and read transactions, balance the target channel bandwidth, and improve bus bandwidth utilization and system transmission efficiency. The functions of bus bridge are verified at the module level, by using Tilelink random test vectors. And by mounting the PCI Express root complex of AXI4 interface, the function of bus bridge is verified at the FPGA system level. The results show that the bus bridge can converse protocol correctly and greatly improve the system bandwidth utilization. The bus bridge is implemented in a SMIC 55 nm CMOS process. The frequency is 714 MHz and the area is 405×405
μ m2 after physical design.-
Key words:
- RISC-V /
- bus bridge /
- TileLink bus /
- AXI4 Bus /
- Pipeline Transmission
-
表 1 PCIE寄存器内存映射
Table 1. PCIE register memory map
寄存器地址 描述 0x000-0x12F PCIe Configuration Space Header 0x130 Bridge Info Register 0x134 Bridge Status and Control Register 0x138 Interrupt Decode Register 0x13C Interrupt Mask Register 0x140 Bus Location Register 0x144 PHY Status/Control Register 0x148 Root Port Status/Control Register 0x14C Root Port MSI Base Register 1 0x150 Root Port MSI Base Register 2 0x154 Root Port Error FIFO Read Register 0x158 Root Port Interrupt FIFO Read Register 1 0x15C Root Port Interrupt FIFO Read Register 2 -
[1] WATERMAN A, LEE Y, PATTERSON D A, et al. The RISC-V instruction set manual, volume I: User-level ISA, version2.1[R]. Berkeley, CA: University of California, Berkeley, 2016. [2] 娄冕, 张海金, 杨靓, 等. 基于Chisel语言的RISC-V处理器设计技术[J]. 微电子学与计算机, 2021, 38(3): 51-55. DOI: 10.19304/j.cnki.issn1000-7180.2021.03.009.LOU M, ZHANG H J, YANG L, et al. Design of RISC-V processor based on chisel[J]. Microelectronics & Computer, 2021, 38(3): 51-55. DOI: 10.19304/j.cnki.issn1000-7180.2021.03.009. [3] SiFive Inc. SiFive TileLink specification[EB/OL]. [2021-12-03]. https://sifive.cdn.prismic.io/sifive/7bef6f5c-ed3a-4712-866a-1a2e0c6b7b13_tilelink_spec_1.8.1.pdf. [4] ARM Co., Ltd. AMBA AXI and ACE protocol specification AXI3, AXI4, and AXI4-Lite ACE and ACE-Lite[EB/OL]. [2021-07-20]. https://developer.arm.com/architectures/system-architectures/amba/specifications. [5] ARM Co., Ltd. AMBATM Specification (Rev 2.0)[EB/OL]. [2021-07-20]. https://developer.arm.com/documentation/ihi0011/a. [6] 付兴飞, 陈胜刚, 曾思, 等. 基于AMBA总线的存储控制器接口设计[J]. 计算机研究与发展, 2015, 52(S2): 119-123. https://www.cnki.com.cn/Article/CJFDTOTAL-JFYZ2015S2018.htmFU X F, CHEN S G, ZENG S, et al. Design and implementation of interface for memory controller based on AMBA bus protocol[J]. Journal of Computer Research and Development, 2015, 52(S2): 119-123. https://www.cnki.com.cn/Article/CJFDTOTAL-JFYZ2015S2018.htm [7] TIDALA N. High performance network on chip using AXI4 protocol interface on an FPGA[C]// 2018 Second International Conference on Electronics, Communication and Aerospace Technology (ICECA). Coimbatore, India: IEEE Press, 2018: 1647-1651. DOI: 10.1109/ICECA.2018.8474664. [8] 张浩, 魏敬和. 高效率PLB2AXI总线桥的设计与验证[J]. 计算机工程, 2020, 46(8): 228-234. DOI: 10.19678/j.issn.1000-3428.0055785.ZHANG H, WEI J H. Design and verification of efficient PLB2AXI bus bridge[J]. Computer Engineering, 2020, 46(8): 228-234. DOI: 10.19678/j.issn.1000-3428.0055785. [9] 焦龙涛, 高欣. 一种低功耗高效率的双向AXI2AHB总线桥设计与实现[J]. 无线电通信技术, 2017, 43(2): 78-82. DOI: 10.3969/j.issn.1003-3114.2017.02.20.JIAO L T, GAO X. Design and verification of a low power and high efficiency AXI2AHB bidirectional bridge[J]. Radio Communications Technology, 2017, 43(2): 78-82. DOI: 10.3969/j.issn.1003-3114.2017.02.20. [10] COOK H, TERPSTRA W, LEE Y. Diplomatic design patterns: A TileLink case study[EB/OL]. [2021-07-20]. https://carrv.github.io/2017/papers/cook-diplomacy-carrv2017.pdf. [11] MCMILLAN K. Modular specification and verification of a cache-coherent interface[C]//2016 Formal Methods in Computer-Aided Design (FMCAD). Mountain View, CA, USA: IEEE Press, 2016: 109-116. DOI: 10.1109/FMCAD.2016.7886668. [12] MATH S S, MANJULA R B, MANVI S S, et al. Data transactions on system-on-chip bus using AXI4 protocol[C]//2011 International Conference on Recent Advancements in Electrical, Electronics and Control Engineering. Sivakasi, India: IEEE Press, 2011: 423-427. DOI: 10.1109/ICONRAEeCE.2011.6129797. [13] LAKHMANI V, ALI N, TRIPATHI V S. AXI compliant DDR3 controller[C]//2010 Second International Conference on Computer Modeling and Simulation. Sanya, China: IEEE Press, 2010: 1-5. DOI: 10.1109/ICCMS.2010.291. [14] ZHANG YJ, KONG X X. Design of Ethernet to optical fiber bridge IP core based on SOPC[C]//201514th International Symposium on Distributed Computing and Applications for Business Engineering and Science (DCABES). Guiyang, China: IEEE Press, 2015: 252-255. DOI: 10.1109/DCABES.2015.70. [15] XIAO F M, LI D S, DU G M, et al. Design of AXI bus based MPSoC on FPGA[C]//20093rd International Conference on Anti-counterfeiting, Security, and Identification in Communication. Hong Kong, China: IEEE Press, 2009: 560-564. DOI: 10.1109/ICASID.2009.5277006. [16] 吴睿振, 杨银堂, 张丽, 等. 一种基于权重与轮询的双层仲裁算法[J]. 电子与信息学报, 2013, 35(12): 3024-3029. DOI: 10.3724/SP.J.1146.2013.00175.WU R Z, YANG Y T, ZHANG L, et al. A two-level arbitration algorithm based on weight and round-robin[J]. Journal of Electronics & Information Technology, 2013, 35(12): 3024-3029. DOI: 10.3724/SP.J.1146.2013.00175. [17] CELIO C, CHIU P F, ASANOVIC K, et al. BROOM: An open-source out-of-order processor with resilient low-voltage operation in 28-nm CMOS[J]. IEEE Micro, 2019, 39(2): 52-60. https://www.computer.org/csdl/magazine/mi/2019/02/08634812/17D45WWzW5v [18] CELIO C, CHIU P F, NIKOLIC B, et al. BOOM v2: An open-source out-of-order RISC-V core: UCB/EECS-2017-157[R]. Berkeley, CA: University of California, Berkeley, 2017. [19] Xilinx Inc. AXI Bridge for PCI Express gen3 subsystem v3.0[EB/OL]. [2021-07-20]. https://www.xilinx.com/support/documentation/ip_documentation/axi_pcie3/v3_0/pg194-axi-bridge-pcie-gen3.pdf. -