(10%) Consider three $n \times n$ matrices, denoted as $A$, $B$, and $C$. Each of these matrices is partitioned into four $n/2 \times n/2$ submatrices. Assuming that $n$ is an exact power of 2, we can guarantee that, for $n \ge 2$, the dimension $n/2$ is an integer. The subdivisions for matrix $A$ yield $A_{11}$, $A_{12}$, $A_{21}$, and $A_{22}$. Similarly, matrix $B$ is decomposed into $B_{11}$, $B_{12}$, $B_{21}$, and $B_{22}$, while matrix $C$ undergoes division into $C_{11}$, $C_{12}$, $C_{21}$, and $C_{22}$. We have the following procedure:

F(A, B)
    n = A.rows
    if n == 1
        c11 = a11 · b11
    else
        C11 = F(A11, B11) + F(A12, B21)
        C12 = F(A11, B12) + F(A12, B22)
        C21 = F(A21, B11) + F(A22, B21)
        C22 = F(A21, B12)
    return C

It is important to note that the procedure for $C_{22}$ is invoked only once and '+' is matrix addition. Let $T(n)$ represent the time required to process two $n \times n$ matrices using this particular procedure. Provide an asymptotic tight bound ($\Theta$) for $T(n)$, assuming that $T(n)$ is a constant for sufficiently small $n$.