time accounting problem (powerpc only?)

[Jörg Sommer]

Hi,

Johannes Berg <johannes at sipsolutions.net> wrote:
>> % diff /proc/interrupts <(sleep 2; cat /proc/interrupts)
>> --- /proc/interrupts    2007-11-23 15:04:06.004846901 +0100
>> +++ /proc/self/fd/11    2007-11-23 15:04:05.952841422 +0100
>> @@ -1,15 +1,15 @@
>>             CPU0       
>> - 25:   18063968   MPIC 1    Level     VIA-PMU
>> + 25:   18064241   MPIC 1    Level     VIA-PMU
>> - 42:    1415066   MPIC 1    Level     keywest i2c
>> - 47:    2075159   MPIC 1    Level     GPIO1 ADB
>> - 48:    6686659   MPIC 1    Level     radeon at pci:0000:00:10.0
>> + 42:    1415084   MPIC 1    Level     keywest i2c
>> + 47:    2075193   MPIC 1    Level     GPIO1 ADB
>> + 48:    6686778   MPIC 1    Level     radeon at pci:0000:00:10.0
>> 
>> I don't know where they come from, but that's the cause of the high IRQ
>> time.
>
> Are you sure about that?

No, I saw the high hardware interrupt value in top and looked at
/proc/interrupts and concluded that those are the evil source of the
high load.

> I'm fairly sure that I always had rather high numbers of interrupt
> here. And the system isn't sluggish or unresponsive as you'd expect if
> the IRQs actually did take 90% of the CPU time!

Okay, so what might be the source?

Doesn't really nobody else see this values? At least, Elimar
Riesebieter has the same problem. He reported on debian-powerpc.

Bye, Jörg.
-- 
Treffen sich zwei Funktionen.
Sagt die eine: „Verschwinde oder ich differenzier' dich!“
Erwidert die andere: „Ätsch, ich bin exponentiell!“