Demystifying ARM SME to Optimize General Matrix Multiplications
2025/12/25 by Chencheng Deng, Deng, Chencheng, Weiling Yang +5 · 11 voices
Computer Science · Engineering · #Low-power high-performance VLSI design #Numerical Methods and Algorithms #Parallel Computing and Optimization Techniques #cs.DC
paper · pdf · doi:10.48550/arxiv.2512.21473
openalex publication_date 2025/12/25 · openalex created_date 2025/12/30 · openalex updated_date 2026/07/28
Abstract
General Matrix Multiplication (GEMM) is a critical kernel in high-performance computing and deep learning. While modern architectures like ARM's Scalable Matrix Extension (SME) introduce dedicated hardware for matrix operations, existing linear algebra libraries fail to fully exploit its potential, particularly for large matrices. This paper presents MpGEMM, an open-source library that leverages key architectural features of SME to optimize GEMM across multiple precisions. Through a systematic characterization of SME, we derive optimization guidelines that inform our design. MpGEMM employs cache-aware partitioning, efficient data packing with on-the-fly transposition, and specialized micro-kernels that utilize multi-vector loads and all available tile registers. Evaluated on an Apple M4 Pro with real-world workloads from DeepSeek and LLaMA, MpGEMM achieves an average speedup of 1.23x over the vendor-optimized Apple Accelerate library and significantly outperforms other open-source alternatives.
Discussions
- Demystifying ARM SME to Optimize General Matrix Multiplications [hn, 88 points, 19 comments]
- ⚡ Hackernews Top story: Demystifying ARM SME to Optimize General Matrix Multiplications [bsky, 1 points, 0 comments]
- #cs.DC Origin | Interest | Match [bsky, 0 points, 0 comments]
- Demystifying ARM SME to Optimize General Matrix Multiplications [bsky, 0 points, 0 comments]
- Demystifying ARM SME to Optimize General Matrix Multiplications #HackerNews https://arxiv.org/abs/2512.21473 [bsky, 0 points, 0 comments]
- Demystifying ARM SME to Optimize General Matrix Multiplications https://arxiv.org/abs/2512.21473 (https://news.ycombinator.com/item?id=46840252) [bsky, 0 points, 0 comments]
- => "Demystifying ARM SME to Optimize General Matrix Multiplications", NUDT, arXiv, Dec 25, 2025 arxiv.org/abs/2512.21473 Apple M4 Pro SME Programmer’s Guide, Ver 1,1, Oct 9 developer.arm.com/documenta [bsky, 0 points, 1 comments]
- https://arxiv.org/abs/2512.21473 ARM SMEの機能と仕組みを詳しく解説する論文です。 一般的な行列乗算の効率的な最適化手法を提案しています。 ARMアーキテクチャでの性能向上に貢献する研究内容です。 [bsky, 0 points, 0 comments]
- Demystifying ARM SME to Optimize General Matrix Multiplications https:// arxiv.org/abs/2512.21473 # arxiv [mastodon, 0 points, 0 comments]
- Demystifying ARM SME to Optimize General Matrix Multiplications https://arxiv.org/abs/2512.21473 [bsky, 0 points, 0 comments]
- Demystifying ARM SME to Optimize General Matrix Multiplications https://arxiv.org/abs/2512.21473 (https://news.ycombinator.com/item?id=46840252) [bsky, 0 points, 0 comments]
Related