Update on DeviceKit

[Rob Taylor]

Holger Macht wrote:
> On Thu 08. May - 14:27:29, Matthew Garrett wrote:
>> On Thu, May 08, 2008 at 03:04:36PM +0200, Holger Macht wrote:
>>> On Thu 08. May - 13:35:11, Matthew Garrett wrote:
>>>> I'm not convinced that fine tuning is something that can be done in a 
>>>> generic way. It's going to be kernel specific, and there's no sense in 
>>>> having it tweakable at a per-user level. Trying to fit it into any sort 
>>>> of dbus interface would basically just be exposing the sysfs interface 
>>>> over dbus, which doesn't sound like a great idea...
>>> We're already doing this quite successful.
>>>
>>> SetCPUFreqPerformance method abstracts all the different sysfs entries of
>>> CPUFreq with one single setting (1 to 100, corresponding to the
>>> performance you get).
>> Well, yes, but what does that mean? You can't rate performance on a 
>> scale of 0-100 - there's multiple factors at play. Is performance a 
>> latency issue? A raw power one? What should the thresholds for ondemand 
>> be? If you're altering multiple factors, then for some workloads a 
>> higher value may yield lower performance. If you're only altering one 
>> factor, you may not be obtaining maximum performance for that power 
>> consumption. Trying to squash this into a linear scale doesn't work.
> 
> You always get more performance for the price of more power
> consumption. And having less performance results in lower power
> consumption and in a non ideal world, you might cope with more latency. At
> least this should be the goal. How can you say it doesn't work? It _does_.
> 

I think I'd prefer to just have a way of setting the governor. In the 
future it would be good to have a kernel interface for setting your 
desired interrupt-wakeup latency, but afaik we don't have this yet.

Thanks,
Rob