Page 90 - 《软件学报》2026年第6期

刘佳玮 等: 面向 RISC-V 架构的深度学习算子测试 2409

[28] Guo QY, Xie XF, Li Y, Zhang XY, Liu Y, Li XH. Audee: Automated testing for deep learning frameworks. In: Proc. of the 35th
IEEE/ACM Int’l Conf. on Automated Software Engineering. Melbourne: IEEE, 2020. 486–498. [doi: 10.1145/3324884.3416571]
[29] Li ZY, Wu JZ, Ling X, Luo TY, Rui ZQ, Wu YJ. The seeds of the FUTURE sprout from history: Fuzzing for unveiling vulnerabilities in
prospective deep-learning libraries. In: Proc. of the 47th IEEE/ACM Int’l Conf. on Software Engineering (ICSE). Ottawa: IEEE, 2020.
1616–1627. [doi: 10.1109/ICSE55347.2025.00132]
[30] Harzevili NS, Mohajer MM, Wei MS, Pham HV, Wang S. History-driven fuzzing for deep learning libraries. ACM Trans. on Software
Engineering and Methodology, 2025, 34(1): 19. [doi: 10.1145/3688838]
[31] Deng YL, Xia CS, Peng HR, Yang CY, Zhang LM. Large language models are zero-shot fuzzers: Fuzzing deep-learning libraries via
large language models. In: Proc. of the 32nd ACM SIGSOFT Int’l Symp. on Software Testing and Analysis. Seattle: ACM, 2023.
423–435. [doi: 10.1145/3597926.3598067]
[32] Deng ZZ, Meng GZ, Chen K, Liu K, Liu T, Xiang L, Chen CY. Differential testing of cross deep learning framework APIs: Revealing
inconsistencies and vulnerabilities. In: Proc. of the 32nd USENIX Conf. on Security Symp. Anaheim: USENIX Association, 2023. 414.
[33] Waterman A, Lee Y, Patterson DA, Asanovic K. The RISC-V instruction set manual, volume I: Base user-level ISA. Technical Report,
UCB/EECS-2011-62, University of California at Berkeley, 2011.
[34] Celio C, Chiu PF, Nikolic B, Patterson DA, Asanović K. BOOMv2: An open-source out-of-order RISC-V core. Technical Report,
UCB/EECS-2017-157, University of California at Berkeley, 2017.
[35] Lopoukhine A, Ficarelli F, Vasiladiotis C, Lydike A, van Delm J, Dutilleul A, Benini L, Verhelst M, Grosser T. A multi-level compiler
backend for accelerated micro-kernels targeting RISC-V ISA extensions. In: Proc. of the 23rd ACM/IEEE Int’l Symp. on Code
Generation and Optimization. Las Vegas: ACM, 2025. 163–178. [doi: 10.1145/3696443.3708952]
[36] Hussain T, Tahir MW, Mushtaq M, Khalid S. Design and benchmarking of a low-cost RISC-V-based high-performance computing
cluster for edge computing. In: Proc. of the 2025 Int’l Conf. on Energy, Power, Environment, Control and Computing (ICEPECC 2025).
Gujrat: IET, 2025. 579–586. [doi: 10.1049/icp.2025.1168]
[37] Titopoulos V, Alexandridis K, Peltekis C, Nicopoulos C, Dimitrakopoulos G. Optimizing structured-sparse matrix multiplication in RISC-V
vector processors. IEEE Trans. on Computers, 2025, 74(4): 1446–1460. [doi: 10.1109/TC.2025.3533083]
[38] Kovač M, Dragić L, Malnar B, Minervini F, Palomar O, Rojas C, Olivieri M, Knezović J, Kovač M. FAUST: Design and implementation
of a pipelined RISC-V vector floating-point unit. Microprocessors and Microsystems, 2023, 97: 104762. [doi: 10.1016/j.micpro.2023.
104762]
[39] Lee Y, Waterman A, Cook H, Zimmer B, Keller B, Puggelli A, Kwak J, Jevtic R, Bailey S, Blagojevic M, Chiu PF, Avizienis R, Richards
B, Bachrach J, Patterson D, Alon E, Nikolic B, Asanovic K. An agile approach to building RISC-V microprocessors. IEEE Micro, 2016,
36(2): 8–20. [doi: 10.1109/MM.2016.11]
[40] Xie XF, Liang Z, Gu P, Basak A, Deng L, Liang L, Hu X, Xie Y. SpaceA: Sparse matrix vector multiplication on processing-in-memory
accelerator. In: Proc. of the 2021 IEEE Int’l Symp. on High-performance Computer Architecture (HPCA). Seoul: IEEE, 2021. 570–583.
[doi: 10.1109/HPCA51647.2021.00055]
[41] Cavalcante M, Schuiki F, Zaruba F, Schaffner M, Benini L. Ara: A 1-GHz+ scalable and energy-efficient RISC-V vector processor with
multiprecision floating-point support in 22-nm FD-SOI. IEEE Trans. on Very Large Scale Integration (VLSI) Systems, 2020, 28(2):
530–543. [doi: 10.1109/TVLSI.2019.2950087]
[42] Gautschi M, Schiavone PD, Traber A, Loi I, Pullini A, Rossi D, Flamand E, Gürkaynak FK, Benini L. Near-threshold RISC-V core with
DSP extensions for scalable IoT endpoint devices. IEEE Trans. on Very Large Scale Integration (VLSI) Systems, 2017, 25(10):
2700–2713. [doi: 10.1109/TVLSI.2017.2654506]
[43] Chen TQ, Moreau T, Jiang ZH, Zheng LM, Yan D, Cowan M, Shen HC, Wang LY, Hu YW, Ceze L, Guestrin C, Krishnamurthy A.
TVM: An automated end-to-end optimizing compiler for deep learning. In: Proc. of the 13th USENIX Conf. on Operating Systems
Design and Implementation. Carlsbad: USENIX Association, 2018. 578–594.
[44] David R, Duke J, Jain A, Reddi Vj, Jeffries N, Li J, Kreeger N, Nappier I, Natraj M, Wang TZ, Warden P, Rhodes R. TensorFlow lite
micro: Embedded machine learning for tinyML systems. Proc. of Machine Learning and Systems, 2021, 3: 800–811.

附中文参考文献
[1] 刘畅, 武延军, 吴敬征, 赵琛. RISC-V 指令集架构研究综述. 软件学报, 2021, 32(12): 3992–4024. http://www.jos.org.cn/1000-9825/
6490.htm [doi: 10.13328/j.cnki.jos.006490]
[3] 韩金池, 王智栋, 马浩, 宋威. Spike-FlexiCAS: 支持缓存架构灵活配置的 RISC-V 处理器模拟器. 软件学报, 2025, 36(9): 3954–3969.