This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

DM6437 device simulator vs eval board (for cache tuning)

Hi,

I am trying to collect cache information (such as Data/Program Cache miss etc.) to improve performance. I have a DaVinci eval board DM6437EVM. I am using the DM6437 device simulator (cycle accurate) to get the Cache data. However, I want to make sure that the simulator has the same SRAM/L2 Cache configuration as my eval board before I can come to any conclusion.

I dug around a bit and found this http://www.ti.com/lit/ug/spru871k/spru871k.pdf 

L2CFG_simulator ---> 0x0100 0002 (64k Cache)

L2CFG_board       ---> 0x0000 0002 (64k Cache)

1.    What I want to know is, is that OK? Does it mean that the simulator results can be trusted?

2.    I ran the simulator with two configurations (zero L2 Cache and 64k L2 cache). The profiler still reports some stats for L2 cache in both cases. What does it mean to have L2 Cache reference when L2 cache size is 0?