Optimization_4x4_9

Copy the contents of file MMult_4x4_8.c into a file named MMult_4x4_9.c and change the contents:

|| from || to || ||<^> -Include(HowToOptimizeGemm/Details/MMult_4x4_8)- ||<^> -Include(HowToOptimizeGemm/Details/MMult_4x4_9)- ||

Change the first lines in the makefile to {{{ OLD := MMult_4x4_8 NEW := MMult_4x4_9 }}}

• make run {{{ octave:3> PlotAll % this will create the plot }}}

This time the performance graph will look something like

ImageLink(http://www.cs.utexas.edu/users/rvdg/HowToOptimizeGemm/Graphs/compare_MMult-4x4-8_MMult-4x4-9.png,http://www.cs.utexas.edu/users/rvdg/HowToOptimizeGemm/Graphs/compare_MMult-4x4-8_MMult-4x4-9.png,width=40%)

From 4x4_8 to 4x4_9 is a subtle change: instead of updating the rows of 4x4 block C a row at a time, we compute them two rows at a time. This sets us up for using vector operations, where we update pairs C( 0,j ) and C( 1,j ) (j =0,...,3) with vector operations.