The idea of ``blocking'' (a.k.a. ``tiling'') is to reorder the execution of the loop nest so that data is not evicted from the cache before it is needed again. E.g.
for (kk = 0; kk < 500; kk += BLKSZ) for (jj = 0; jj < 500; jj += BLKSZ) for (i = 0; i < 500; i++) for (k = kk; k < min(kk+BLKSZ,500); k++){ r = A[i][k]; for (j = jj; j < min(jj+BLKSZ, 500); j++) C[i][j] += r * B[k][j]; }