On Tue, Feb 28, 2012 at 3:33 PM, Rob Lee rob.lee@linaro.org wrote:
I brought this topic up internally and Jon suggested that the 'usage' statistics that are reported in sysfs should also reflect failed versus successful C-state transitions, which is a great idea. This could simply be achieved by renaming the current 'usage' count to something like 'transitions_attempted' and then conditionally increment a new counter within the 'if (entered_state >= 0)' block, perhaps named, 'transition_succeeded'.
This way the old 'usage' count paradigm is as accurate as the new time-keeping code. Being able to easily observe which C-state tend to fail the most would be invaluable in tuning idle policy for maximum effectiveness.
Thoughts?
Sounds reasonable. In some cases it may be helpful to track state demotion as well. Since I'm still a noob and wearing my submission training wheels, I'm trying to minimize things that fall outside of this basic consolidation effort for this patch series. But I added Jon's suggestion to this cpuidle page which contains future cpuidle items to consider adding: https://wiki.linaro.org/WorkingGroups/PowerManagement/Doc/CPUIdle#Track_both...
Yeah, I don't want to feature-bloat your submission more than necessary. I'm happy for the usage counter stuff to get tackled at a later date, but you're still on board for setting last_residency to zero in this series, right?
Regards, Mike
Regards, Mike
Regards, Mike