ARM Matrix Multiplication (NEON Edition)

说明与用途

使用 ARM NEON 汇编实现矩阵乘法的指南，展示手动优化和性能测量。

问题

Implement fast matrix multiplication on ARM CPUs using NEON assembly instead of relying on GPU or high-level libraries.

结果

Hand‑written NEON assembly slightly outperforms the C compiler but remains a small fraction of the M2’s theoretical peak. SIMD vectorization yields modest gains, highlighting the need for more advanced optimizations (e.g., SME/SVE2).

工作流

• Write reference C matrix multiplication
• Compile with -O2 and benchmark
• Translate naive C loops to ARM assembly
• Benchmark assembly version
• Add SIMD vectorization (load 4 B values, broadcast A, use fmadd)
• Benchmark SIMD version

证据

At -O2 this runs at 2.09 Gflops, about 1.9% of our theoretical maximum.

来源

alright we have 1.98 Gflops (or about 1.8% of the theoretical max)... the c compiler is smarter than me.

来源