Processor Specialization

Processor Specialization

Enabling agile development of domain-specific processors through unified interfaces and automated integration.

Our research enables agile development of domain-specific processors through:

Unified ISAX Interface

Cross-platform portability across RISC-V ecosystems. Our unified interface allows ISAX designs to be seamlessly deployed on different processor implementations.

Automated Integration

Seamless ISAX integration with Chipyard and PULP platforms. Our tools automate the process of integrating custom instructions into existing processor frameworks.

ISAX-Level Parallelism

Support for multiple concurrent ISAXs, boosting performance in a new dimension.


Related Publications

  • Customizing Scalable Instructions via Multi-Level E-Graph Anti-Unification
    Yitian Sun, Youwei Xiao, Yuyang Zou, Chenyun Yin, Yun Liang
    ASPLOS 2027
    DOI
  • Aquas: Enhancing Domain Specialization through Holistic Hardware-Software Co-Optimization based on MLIR
    Yuyang Zou, Youwei Xiao, Yansong Xu, Chenyun Yin, Yuhao Luo, Yitian Sun, Ruifan Xu, Renze Chen, Yun Liang
    ICCAD 2026
    DOI
  • Finding Reusable Instructions via E-Graph Anti-Unification
    Youwei Xiao, Chenyun Yin, Yitian Sun, Yuyang Zou, Yun Liang
    ASPLOS 2026 Best Paper Award
    DOI Code
  • APS: Open-Source Hardware-Software Co-Design Framework for Agile Processor Specialization
    Youwei Xiao, Yuyang Zou, Yansong Xu, Yuhao Luo, Yitian Sun, Chenyun Yin, Ruifan Xu, Renze Chen, Yun Liang
    ICCAD 2025 Invited Paper
    DOI
  • POPA: Expressing High and Portable Performance across Spatial and Vector Architectures for Tensor Computations
    Xiaochen Hao, Hongbo Rong, Mingzhe Zhang, Ce Sun, Hong Jiang, Yun Liang
    FPGA 2024
    DOI Code

← Back to all research areas