Hi Eduardo, > > > > > Patch 2/4: Introduce fair_share governor > > This throttles the cooling_devices according to their > > weights (and hence the name; Suggestion are welcome :-). > > The weights in turn describe the effectiveness of a > > particular cooling device in cooling a thermal zone. > > For the purpose you mentioned the name sounds Ok. > > On the other hand, Now I got a bit confused on this strategy. > > Do you still keep the trip to cooling device binding constraints? > Does it make sense to have this binding coming from the pdata > description as well? Yes it makes sense. Now that we can have multiple active and passive trip points, providing 'per-trip point binding' tends to become complex [1]. > > * Add more protection and tidy up the existing ones > > * Expose the weights and cooling devices through sysfs (Read-Only) > > * Remove all throttling related code(if we all agree) from thermal_sys.c > > I do. +1 Thank you. > > * If we all agree, use step_wise and remove linear_throttle from thermal_sys.c > > Well, I guess I need first to understand the difference between those two. > For instance, does it make sense to have a separate file for linear_throttle? There is no difference. I remember specifying it in one of the comments in the patches. I was not sure about our opinions on separating throttle logic from thermal_sys.c (even for simple linear throttle). That's why kept both of them here. Now that we agree, I will remove throttling code inside thermal_sys.c, and keep only step_wise_throttle. I hope this will make things clear, and I will be able to split patches in a better way and make them smaller. Will do this in my next patch set. > > * Find a way to provide platform data so that we can map cooling devices > > for a trip point in a thermal zone. > > Ohh ok.. Indeed, we need a way to describe the mappings and bindings. > I am not sure how that goes into ACPI, but I guess it comes from firmware. > > On non-ACPI world we still need to have that mapped somehow. [1] I thought through this, and (to me) it looks like a 3D mapping between thermal zones -> cooling devices -> trip points of a thermal zone. So, I simplified it and ended up with this 'weight' approach for fair_share This can be a start, and we can take it forward from here.. I think this is a separate topic in itself, which would require one more complete mail chain :-) > > Your weighting patch is at least one attempt to start doing this mapping. > I guess we need more brainstorming here.. Yes. Agree with you. Thanks, Durga -- To unsubscribe from this list: send the line "unsubscribe linux-acpi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
