miknoj / matmul.c

High-Performance Matrix Multiplication on CPUs

Important! The code should be compiled with Clang. GCC causes the program to run 1.5 to 2 times slower on my machine. Please don’t expect peak performance without fine-tuning the hyperparameters, such as the number of threads, kernel and block sizes, unless you are running it on a Ryzen 7700(X). More on this in the tutorial.

Key Features

• Simple and scalable C code (<150 LOC)
• Step by step tutorial
• Targets x86 processors with AVX and FMA instructions (=all modern Intel Core and AMD Ryzen CPUs)
• Faster than OpenBLAS when fine-tuned for Ryzen 7700
• Efficiently parallelized with just 3 lines of OpenMP directives
• Follows the BLIS design
• Works for arbitrary matrix sizes
• Intuitive API void matmul(float* A, float* B, float* C, const int M, const int N, const int K)

How to use

For quick testing, fine-tuning, and prototyping, use the standalone file matmul.c in the main folder:

clang-17 -O2 -mno-avx512f -fopenmp -march=native matmul.c -o matmul.out && ./matmul.out

To verify the numerial accuracy of the implementation, add -DTEST:

clang-17 -O2 -mno-avx512f -fopenmp -march=native -DTEST matmul.c -o matmul.out && ./matmul.out

Performance

Tested on:

• CPU: Ryzen 7 7700 8 Cores, 16 Threads
• RAM: 32GB DDR5 6000 MHz CL36
• Numpy 1.26.4
• Compiler: clang-17
• Compiler flags: -O2 -mno-avx512f -march=native
• OS: Ubuntu 22.04.4 LTS

To reproduce the results, run:

python benchmark_numpy.py

clang-17 -O2 -mno-avx512f -march=native -fopenmp benchmark.c -o benchmark.out && ./benchmark.out

python plot_benchmark.py