Scholarlink

<p>Achieving high performance for matrix multiplication (GEMM) remains critical in high-performance computing (HPC) and machine learning (ML). As core computational components, micro-kernels determine GEMM efficiency but face challenges in development cost and hardware adaptability—especially on modern General-Purpose DSPs (GPDSPs) with VLIW and SIMD architectures. Existing automatic micro-kernel generation methods on DSP platforms either suffer from suboptimal performance due to general compiler backend limitations or face challenges in handling complex VLIW instruction scheduling and cross-platform adaptability. This paper proposes MAGGIE, an MLIR-based automatic micro-kernel generation approach for GPDSPs that uses a progressive lowering design at two levels: generic optimization (i.e., algorithmic transformations and intermediate code optimization) and backend specialization (i.e., DSP-specific instruction scheduling and memory access fusion). Experimental results on the FT-M7032 DSP platform demonstrate that automatically generated micro-kernels achieve a peak efficiency of 99.2%, comparable to that of expert-crafted assembly code. This method significantly reduces development costs while improving performance portability.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

High-performance matrix multiplication micro-kernel generation for GPDSPs via MLIR progressive lowering

  • 期刊论文

  • 发表日期：

    2026年5月18日

查看全文
  • Jingyu Liu,
  • Kainan Yu,
  • Peng Zhang,
  • Chun Huang,
  • Jianbin Fang

摘要

Achieving high performance for matrix multiplication (GEMM) remains critical in high-performance computing (HPC) and machine learning (ML). As core computational components, micro-kernels determine GEMM efficiency but face challenges in development cost and hardware adaptability—especially on modern General-Purpose DSPs (GPDSPs) with VLIW and SIMD architectures. Existing automatic micro-kernel generation methods on DSP platforms either suffer from suboptimal performance due to general compiler backend limitations or face challenges in handling complex VLIW instruction scheduling and cross-platform adaptability. This paper proposes MAGGIE, an MLIR-based automatic micro-kernel generation approach for GPDSPs that uses a progressive lowering design at two levels: generic optimization (i.e., algorithmic transformations and intermediate code optimization) and backend specialization (i.e., DSP-specific instruction scheduling and memory access fusion). Experimental results on the FT-M7032 DSP platform demonstrate that automatically generated micro-kernels achieve a peak efficiency of 99.2%, comparable to that of expert-crafted assembly code. This method significantly reduces development costs while improving performance portability.