Possible Bug on find_energy_efficient_cpu.
I was studying your code about EAS on CFS and I saw something that made me thinking.
In your code in fair.c ::find_energy_efficient_cpu() at line 6293 there is this instruction:
base_energy_pd = compute_energy(p, -1, pd);
So in base_energy_pd there is the energy consumed by the core without p.
After this there is at line 6317
prev_delta = compute_energy(p, prev_cpu, pd);
That compute the energy with p on core prev_cpu.
My question is: Is safe do this operation without checking that OPP is the same? Could p on prev_cpu rising the OPP of the entire pd?
Thanks a lot!
To upload designs, you'll need to enable LFS and have an admin enable hashed storage. More information