On Tuesday 26 October 2010 15:17:43 Thomas Renninger wrote:
> On Tuesday 26 October 2010 13:54:22 Ingo Molnar wrote:
> >
> > * Thomas Renninger <trenn@xxxxxxx> wrote:
...
> If cpuidle is not registered, the events you get are arch specific.
> I mean they are arch specific anyway, but with cpuidle you can
> build up an arch independent userspace framework nicely by looking
> up name/desc/power/... of an cpu_idle event in cpuidle sysfs as
> described above.
About cpuidle and cpu_idle events:
There is some oddness that:
arch specific code which registers for cpuidle has to
throw the cpu_idle enter sleep state X event
and the generic cpuidle framework triggers the "exit" event.
So as there are only cpu_idle events in drivers/idle/intel_idle.c,
but not in drivers/acpi/processor_idle.c, I expect that processor.ko
idle driver is broken and only exit states are sent.
Ideally all cpuidle events should be thrown in cpuidle.c like this:
trace_processor_idle(target_state, smp_processor_id());
dev->last_residency = target_state->enter(dev, target_state);
trace_processor_idle(PWR_EVENT_EXIT, smp_processor_id());
My patches do not touch this behavior. If, it was broken before.
I'll look at it separately.
Thomas
--
To unsubscribe from this list: send the line "unsubscribe linux-trace-users" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
|