Arjan van de Ven <[EMAIL PROTECTED]> writes: > John Fremlin wrote: > >> For a very simple example: suppose the CPU enters state C1. Then >> cur_usage[1]=1. Now suppose that it stays there for a few >> milliseconds. It will automatically fall down to C2. > > > actually.. it will not unless the OS kernel wakes up, executes code > and then selects C2. Even if the HW would change C-states under you, > the counters wouldn't be updated to reflect that because it's the > kernel that does that counting pretty much ;)
OK. I thought from this diagram that the hw did change state automatically, but now I see that I misread it. http://www.intel.com/technology/itj/2006/volume10issue02/art03_Power_and_Thermal_Management/figures/figure_1_lg.gif But your explanation definitely scotches my theory, leaving the problem of many unexplained wake-ups. >> So >> cur_usage[2]=1. Then there is a wake-up and it goes back to C0. Your >> calculation will say there were two wake-ups. > > because there were two... So what were they!? They are not explained by adding up the known wake-up causes! _______________________________________________ Power mailing list [email protected] http://www.bughost.org/mailman/listinfo/power
