On 01/28/2016 03:11 PM, Viresh Kumar wrote: > On 28-01-16, 15:06, Shilpasri G Bhat wrote: >> No these stats are not per-policy. They are per-chip. The throttle event is >> common for all cores in the chip. > > How do you define a chip? And how is it different then the group of > CPUs represented by the policy ? >
Chip is a group of policies. Hmm yes I see your point. We anyways maintain frequency stats which is per-policy. We might as well have throttle stats exported per-policy which points to per-chip data.
