Bill Holler wrote: > Deep cstate avoidance based on the number of non-idle cpus in the > system does not play well with powertop! Odd. > May be the dtrace probes with powertop causes all cpus to be busy at > the same times? > > > When powertop is not running, non-idle cpu count throttling of > deep c-states does not effect idle systems. >
How is a cpu considered to be non-idle? we need to figure out the workload is light or heavy on this cpu. Consider my patch for 5444, it makes powertop happy, :-) -Aubrey
