Application of asynchronous data bypass to increase performance of in-order superscalar RISC-V processor core

Software and Hardware of Computer, Network, Telecommunication, Control, and Measurement Systems
Authors:
Usoltsev M.D. Borisovskii F.K. Budanov D.O.
Abstract:

RISC-V is an open-source architecture that does not require license fees from pro-cessor core vendors. This allowed RISC-V to become widely distributed first in the micro-controller-class processor core market and subsequently in the market of higher-performance devices. Currently RISC-V processor cores with in-order execution, which combine mid-range performance, compactness, and energy efficiency are attracting the most interest from customers. This makes the task of continuing to improve them a highly relevant. This paper explores the application of a microarchitectural modification, asynchronous data bypass, to increase the performance of in-order superscalar RISC-V processor core. Asynchronous data bypass has been made for two pairs of functional blocks: between arithmetic-logic unit (ALU) and ALU, between ALU and branch unit. The main obstacle of implementing asynchronous data bypass is the creation of an extremely long logical path, which leads to violations of timing constraints. To overcome this obstacle, an approach was proposed that involves completely excluding a subset of instructions from asynchronous bypass and separating the logic of the remaining instructions into groups based on the length of the logical path. This approach made it possible to keep the logical path of asynchronous bypass no longer than the longest path of a single execution unit. The results showed that the use of asynchronous bypass can significantly increase the performance of in-order superscalar processor core. At the same time, the area overhead is very low.

Key words:
processor coresuperscalarityin-orderasynchronous data bypasslogical path
  • References

    1. Chaver D., Harris S., Pinuel L., Kindgren O., Kakakhel Z., Owen C. The RISC-V FPGA (RVfpga) Teaching Package. IEEE Access, 2026, Vol. 14, Pp. 18455–18475. DOI: 10.1109/ACCESS.2026.3658743

    2. Sharma M., Bhatnagar E., Puri K., Mitra A., Agarwal J. A survey of RISC-V CPU for IoT applicati-ons. Proceedings of the International Conference on Innovative Computing & Communication (ICICC), 2022, Pp. 1–4. DOI: 10.2139/ssrn.4033491

    3. Hassan Q.F., Sagahyroon A. RISC-V: A comprehensive overview of an emerging ISA for the AI-IoT era. In: Advances in the Internet of Things (ed. Q.F. Hassan), 2025, Pp. 244–284. DOI: 10.1201/978100350-6638-15

    4. Liu Y., Ye K., Xu C.Z. Performance evaluation of various RISC processor systems: A case study on ARM, MIPS and RISC-V. Cloud Computing – CLOUD 2021, 2022, Vol. 12989, Pp. 61–74. DOI: 10.1007/978-3-030-96326-2_5

    5. Gao J., Zhang J. Research and Design of RISC-V Four-Stage Out-of-Order Execution Processor. 2022 IEEE 16th International Conference on Solid-State & Integrated Circuit Technology (ICSICT), 2022, Pp. 1–3. DOI: 10.1109/ICSICT55466.2022.9963329

    6. Fu Z., Tedeschi R., Ottavi G., Wistoff N., Fuguet C., Rossi D., Benini L. Ramping up open-source RISC-V cores: Assessing the energy efficiency of superscalar, out-of-order execution. Proceedings of the 22nd ACM International Conference on Computing Frontiers, 2025, Pp. 12–20. DOI: 10.1145/3719276.3725186

    7. Chatzopoulos O., Papadimitriou G., Wong W.S., Gizopoulos D. Energy Efficiency of Out-of-Order CPUs: Comparative Study and Microarchitectural Hotspot Characterization of RISC-V Designs. 2023 IEEE International Symposium on Workload Characterization (IISWC), 2023, Pp. 216–220. DOI: 10.1109/IISWC59245.2023.00032

    8. Cui E., Li T., Wei Q. RISC-V Instruction Set Architecture Extensions: A Survey. IEEE Access, 2023, Vol. 11, Pp. 24696–24711. DOI: 10.1109/ACCESS.2023.3246491

    9. Schiavone P.D., Conti F., Rossi D., Gautschi M., Pullini A., Flamand E. Slow and steady wins the race? A comparison of ultra-low-power RISC-V cores for Internet-of-Things applications. 2017 27th International Symposium on Power and Timing Modeling, Optimization and Simulation (PATMOS), 2017, Pp. 1–8. DOI: 10.1109/PATMOS.2017.8106976

    10. Wheeler B. WD Rolls Its Own RISC-V Core: Western Digital Opens Superscalar Swerv CPU Source to Others. The Linley Group: Microprocessor Report, 2019, Pp. 1–3. [online] Available: https://documents.westerndigital.com/content/dam/doc-library/en_us/assets/public/western-digital/collateral/analyst-report/WD_Rolls_Its_Own_RISC-V_Core.pdf

    11. Western Digital. RISC-V and Open Source Hardware Address New Compute Requirements. [online] https://documents.westerndigital.com/content/dam/doc-library/en_us/assets/public/western-digital/collateral/tech-brief/tech-brief-western-digital-risc-v.pdf (Accessed 28.07.2026)

    12. Petrisko D., Gilani F., Wyse M., Jung D.C., Davidson S., Gao P. BlackParrot: An Agile Open-Source RISC-V Multicore for Accelerator SoCs. IEEE Micro, 2020, Vol. 40, No. 4, Pp. 93–102. DOI: 10.1109/MM.2020.2996145

    13. Gonzalez A., Zhao J., Korpan B., Genc H., Schmidt C., Wright J. et al. A 16mm2 106.1 GOPS/W Heterogeneous RISC-V Multi-Core Multi-Accelerator SoC in Low-Power 22nm FinFET. ESSCIRC 2021 – IEEE 47th European Solid State Circuits Conference (ESSCIRC), 2021, Pp. 259–262. DOI: 10.1109/ESSCIRC53450.2021.9567768

    14. Numan Ince M., Ledet J., Gunay M. Building An Open Source Linux Computing System On RISC-V. 2019 1st International Informatics and Software Engineering Conference (UBMYK), 2019, Pp. 1–4. DOI: 10.1109/UBMYK48245.2019.8965559

    15. Gookyi D.A.N., Ryoo K. Selecting a Synthesizable RISC-V Processor Core for Low-cost Hardware Devices. Journal of Information Processing Systems, 2019, Vol. 15, No. 6, Pp. 1406–1421. DOI: 10.3745/JIPS.03.0129

    16. Dörflinger A., Albers M., Kleinbeck B., Guan Y., Michalik H., Klink R., Blochwitz C., Nechi A., Berekovic M. A comparative survey of open-source application-class RISC-V processor implementations. Proceedings of the 18th ACM International Conference on Computing Frontiers, 2021, Pp. 12–20. DOI: 10.1145/3457388.345865

    17. Li Y., Yu Z., Ji W. Design and Performance Optimization of a Risc-V Processor with Five-Stage Pipeline. 2025 IEEE 6th International Seminar on Artificial Intelligence, Networking and Information Technology (AINIT), 2025, Pp. 1–5. DOI: 10.1109/AINIT65432.2025.11035005

    18. Kra Y., Shoshan Y., Rudin Y., Teman A. HAMSA-DI: A Low-Power Dual-Issue RISC-V Core Targeting Energy-Efficient Embedded Systems. IEEE Transactions on Circuits and Systems I: Regular Papers, 2024, Vol. 71, No. 1, Pp. 223–236. DOI: 10.1109/TCSI.2023.3323425

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License
Previous articleNext article