Floating-point matrix multiplication implementation (arbitrary precision) - View it on GitHub