Performance of blocked version

Blocking factor	Execn time	MFLOPS
8	3.815	70.4
16	2.784	96.4
32	2.283	117.6
40	2.193	122.4
48	2.253	119.1
56	2.473	108.5
64	3.404	78.9
72	5.608	47.9
80	5.578	48.1
88	5.808	46.2
96	5.928	45.3
104	6.309	42.5
112	5.778	46.5