Hello,
I would like to deploy a ML model on a Arm cortex-A core (Neoverse N1) targeting bare metal. I need bare metal in order to simulate the inference runtime in a execution testbench in ModelSim. Therefore I would like to use microTVM but I am not sure if it supports or is able to compile bare metal code for such architectures?
For instance, this line of code is used for targeting Cortex-M core tvm.target.target.create(c -keys=arm_cpu -mcpu=cortex-m7 -link-params -model=stm32.. -runtime-c -system-lib=l )
Can I replace -mcpu with another key option like cortex-a?
Thank you very much in advance for your help!