[wellylug] ubuntu load average

[Pete Black]

Well pardon me for relating my experience and trying to help.

So this:

http://forums.whirlpool.net.au/forum-replies-archive.cfm/385948.html

and this:

https://lists.ubuntu.com/archives/ubuntu-devel/2006-November/022738.html

and this:

http://www.linuxquestions.org/questions/linux-laptop-and-netbook-25/powertop-reporting-a-huge-number-of-acpi-interrupts-597349/

and the bunch of other stuff you get from google when googling 'ubuntu 
rtc interrupt idle load' is obviously irrelevant to this discussion?

*shrug*

-Pete




> Pete Black <pete at marchingcubes.com> writes:
>
>   
>> I've personally seen rtc interrupt wierdness causing idle load
>> problems on one of my AMD machines, though that machine has a
>> completely screwy APIC/ACPI implementation too.
>>     
>
> OK, good.  How did it cause that?
>
>   
>> If its not the issue, its not the issue, but its something to check
>> into. if /proc/interrupts shows thousands or millions of rtc
>> interrupts, while all the other interrupt counts are low, you don't
>> think that might be an indication of a problem in this area?
>>     
>
> Um, no.  At 1000, or even 250, interrupts a second you can have
> significantly more RTC interrupts shown than any other source, given
> they typically average somewhere between 0 and 100 a second.
>
> I mean, sure, check into it, but it seems highly improbably that it is
> related, and so far no one — not even you — has actually pointed to a
> connection between RTC interrupt problems and high idle load.
>
> *shrug*
>
> Regards,
>         Daniel
>
>
>