more cache misses than on raspberry pi 3 model B

[jiez]

Hi,

Sorry this is not an issue but a question. I am running a test program on VIM3L. It is slower than running on raspberry pi 3 model B, i.e. needs more cycles. I used the same binary. This is strange. I tried to dig deeper. I found when running on VIM3L, it has 60% more L1 data cache read misses than running on raspberry pi 3 model B. The L1 data cache is 32KB per core on VIM3L, and is 16KB per core on raspberry pi 3 model B. So it is rather strange. I cannot share the test program publicly. But do you have any experience on this? I'm running linux kernel 4.9.206 on VIM3L, and 5.4.0 on raspberry pi 3 model B. Thanks.

[jiez]

The cause of this issue was found in the test program.