while having a discussion over the newest udev 084 in testing, i did a reboot after weeks and guess what i found? p4_clockmod was probed which depends on speedstep_lib and freq_table. being familiar them from embedded machines at work i instantly tried powernowd against it.
i loaded cpufreq_userspace and started up powernowd. neither of them did complain and powernowd told me that it detected a cpu capable of stepping down from its maximum 2.4Ghz to 300Mhz, wow. the next breath i was able to catch was when the cpu actually *was* running on only 300Mhz... so it seemed. note that /proc/cpuinfo was not updated but the system actually felt way slower when i forced the speed to 300Mhz which made me thinking...
anyone else experienced this?
I recognize that while theory and practice are, in theory, the same, they are, in practice, different. -Mark Mitchell
I haven't messed around with cpufreq myself, but doesn't it rely on changing the multiplier to ramp the processor down? You'd have to have an unlocked chip (such as a mobile version, or a modded one) in order for it to actually change...unless it was adjusting the clock frequency directly. Weird results either way!