SDK performance monitoring not catching cache misses
I was trying to use the performance monitoring tool to monitor cache misses when I ran my own code. The code was a simple Z = Ax + B where A, B, and Z are 2D arrays and x is a constant. A, B, and Z are all 1000 x 1000 char arrays. I'm using malloc to declare the arrays and my heap is set to 4MB. This code should be causing cache misses but when I check using performance monitoring, my numbers don't make sense (i.e. it says my miss rate is .04%). I'm using a Zynq Ultrascale + ZU106 board. I've been following this:https://xilinx-wiki.atlassian.net/wiki/spaces/A/pages/18842421/Zynq+UltraScale+MPSoC+-+System+Performance+Modelling
Why isn't it monitoring the cache misses properly? I'm not quite sure what I'm doing wrong and for the project, I need to be able to monitor cache misses.