High [extra timer interrupt] count in powertop since 2.6.36

High [extra timer interrupt] count in powertop since 2.6.36

[Ian Kumlien]

Hi,

I have been searching the list for anyone else seeing this,
but i haven't found anything yet - Any ideas on how to debug?

>From powertop:
Top causes for wakeups:
   84.7% (1673.3)   [extra timer interrupt]

Thats a little bit high in my opinion, for some reason it seems to be
ignored since the actual number of wakeups seems to be ~367 or so.

Ideas? Suggestions?

Re: High [extra timer interrupt] count in powertop since 2.6.36

[Jeremy Fitzhardinge]

 On 10/30/2010 07:52 PM, Ian Kumlien wrote:
> Hi,
>
> I have been searching the list for anyone else seeing this,
> but i haven't found anything yet - Any ideas on how to debug?
>
> From powertop:
> Top causes for wakeups:
>   84.7% (1673.3)   [extra timer interrupt]
>
> Thats a little bit high in my opinion, for some reason it seems to be
> ignored since the actual number of wakeups seems to be ~367 or so.
>
> Ideas? Suggestions?

I'm not sure, but I'm seeing something similar.

    J

Re: High [extra timer interrupt] count in powertop since 2.6.36

[Ian Kumlien]

On Mon, 1 Nov 2010, Jeremy Fitzhardinge wrote:

> On 10/30/2010 07:52 PM, Ian Kumlien wrote:
>> Hi,
>>
>> I have been searching the list for anyone else seeing this,
>> but i haven't found anything yet - Any ideas on how to debug?
>>
>> From powertop:
>> Top causes for wakeups:
>>   84.7% (1673.3)   [extra timer interrupt]
>>
>> Thats a little bit high in my opinion, for some reason it seems to be
>> ignored since the actual number of wakeups seems to be ~367 or so.
>>
>> Ideas? Suggestions?
>
> I'm not sure, but I'm seeing something similar.

using powertop: 1.13

< Detailed C-state information is not P-states (frequencies)
                                         3.21 Ghz     6.0%
                                         2.40 Ghz     0.0%
                                         1.60 Ghz     0.1%
                                          800 Mhz    93.9%


Wakeups-from-idle per second : 441.8    interval: 10.0s 
no ACPI power usage estimate available

Top causes for wakeups:
   84.3% (2132.0)   [extra timer interrupt]
    8.1% (205.4)   qemu-system-x86
    2.4% ( 61.0)   [eth1] <interrupt>
    1.2% ( 31.1)   [eth0] <interrupt>
    1.2% ( 29.3)   java
    1.0% ( 26.3)   [ahci] <interrupt>
    0.5% ( 13.0)   [sky2@pci:0000:05:00.0] <interrupt>
    0.5% ( 12.0)   kworker/u:2
...

Has any output been changed, is this just a missenterpetation done by powertop?

CPU: AMD Phenom(tm) II X6 1090T Processor

>    J
>
>

Re: High [extra timer interrupt] count in powertop since 2.6.36

[Maciej Rutecki]

On niedziela, 31 października 2010 o 01:52:08 Ian Kumlien wrote:
> Hi,
> 
> I have been searching the list for anyone else seeing this,
> but i haven't found anything yet - Any ideas on how to debug?
> 
> From powertop:
> Top causes for wakeups:
>    84.7% (1673.3)   [extra timer interrupt]
> 
> Thats a little bit high in my opinion, for some reason it seems to be
> ignored since the actual number of wakeups seems to be ~367 or so.
> 
> Ideas? Suggestions?

I created a Bugzilla entry at 
https://bugzilla.kernel.org/show_bug.cgi?id=22272
for your bug report, please add your address to the CC list in there, thanks!

-- 
Maciej Rutecki
http://www.maciek.unixy.pl