On 4/20/2011 11:08 PM, John Mathew wrote:
> Intel hyperthreaded cores does not exactly behave like if it was 2 real cores.
> - Actual HW cstate is roughly minimal cstate of the 2 threads.
> As most of the resources are shared, HW resources are only really
> powered
> down when both threads are idle.
> - Interrupts always wakes up the 2 threads. i.e. terminate the 2 mwait.
> So in case of an hyperthreaded core a wakeup event is affecting power only
> when
> both the threads are asleep. A wakeup event that occurs when one of
> the thread is active should not be considered as a wakeup event.
while your reasoning might be right.... (I'll argue that but hey...)
your code is not unfortunately ;-(.
}
> + if (strstr(line, "ht")) {
> + hyper_threading = 1;
> + }
the "ht" flag only means that the MSRs that report the number of
siblings is valid. It does not mean that you actually have siblings.
> + if (cpu& 1){
and this "it's the last bit" is not correct behavior; first of all,
that's only valid for APIC IDs not Linux CPU numbers, and only IF you
have exactly 2 siblings,
which you don't check. Reality is that you must use the topology
information exported by the kernel if you want to use this.
_______________________________________________
Power mailing list
[email protected]
https://bughost.org/mailman/listinfo/power
