代码

#pragma GCC optimize("Ofast", "unroll-loops", "omit-frame-pointer", "inline",  \
    "-ftree-vectorize", "-fopt-info-vec-all")
#pragma GCC option("arch=native", "tune=native", "no-zero-upper")
#pragma GCC target("sse,sse2,sse3,sse4.1,sse4.2,popcnt,abm,mmx,avx,avx2")
#define idx(i, j) ((i) * n + (j))

void matrix_multiply(int n, const double *A, const double *B, double *C) {
	for (int k = 0; k < n; ++k) {
        for (int i = 0; i < n; ++i) {
            double r = A[idx(i,k)];
            for (int j = 0; j < n; j += 4) {
                C[idx(i,j)] += r * B[idx(k,j)];
                C[idx(i,j+1)] += r * B[idx(k,j+1)];
                C[idx(i,j+2)] += r * B[idx(k,j+2)];
                C[idx(i,j+3)] += r * B[idx(k,j+3)];
            }
        }
    }
}

评测结果