Is it possible to load the A72 cortex with more computation work compared to A53 cortex? Specially during multithreading can i generate more cores on A72 cortex as it is significantly faster?

While compiling there is a flag to select the appropriate cortex (-mtune=cortex-a72.cortex-a53) but it only generates appropriate cortex code.

Any way to control these cortex based on the application?  

Are you talking about CPU affinity at run-time, at the process level you can use taskset or syscalls to bind a process to a core... or do you mean on the thread level?

@Xalius

I meant on the thread level. If i could assign more threads to A72 core and regulate the threads processed by A53.

Thanks

I  think it will work only in async impl. mean one piece of code independantly run a72 and 'supervisors' on a53 or even lower core. Like docker application worker nodes on faster nodes  and agents on slow ones.