Re: 20000+ wake-ups/second in 2.6.24. Bug?

[Mark Lord <lkml@rtr.ca>]

Arjan van de Ven wrote:
> On Mon, 04 Feb 2008 12:29:03 -0500
> Mark Lord <lkml@rtr.ca> wrote:
> 
>> re:  http://bugzilla.kernel.org/show_bug.cgi?id=9489
>>
>> This just happened here again.  Or at least I finally noticed that
>> the fan on my notebook seemed to be running hard for much longer
>> than usual.  :)
>>
>> Powertop showed 2.6.24-final running with 10000-36000 wakeups/sec,
>> with *nothing* significant running:  top showed 97+% idle on both
>> cores.
>>
>> -               Device: Errors: Correctable- Non-Fatal- Fatal+
>> Unsupported-
>> +               Device: Errors: Correctable+ Non-Fatal+ Fatal+
>> Unsupported+ Device: RlxdOrd- ExtTag- PhantFunc- AuxPwr- NoSnoop-
>>                 Device: MaxPayload 128 bytes, MaxReadReq 128 bytes
>>                 Link: Supported Speed 2.5Gb/s, Width x1, ASPM L0s L1,
>> Port 1 @@ -101,12 +101,12 @@
>>         Latency: 0, Cache Line Size: 64 bytes
>>         Bus: primary=00, secondary=0c, subordinate=0c, sec-latency=0
>>         Memory behind bridge: efc00000-efcfffff
>> -       Secondary status: 66MHz- FastB2B- ParErr- DEVSEL=fast
>>> TAbort- <TAbort- <MAbort- <SERR- <PERR-
>> +       Secondary status: 66MHz- FastB2B- ParErr- DEVSEL=fast
>>> TAbort- <TAbort- <MAbort+ <SERR- <PERR- BridgeCtl: Parity- SERR+
>>> NoISA- VGA- MAbort- >Reset- FastB2B-
> 
> this shows you're having various types of really bad things going on, like PCI
> master aborts and the like. Those would certainly be a factor in waking the cpu up;
> they're basically hardware exceptions, and I can totally believe (would need to find out
> from hw guys how this works in practice) that this sort of serious error would keep the
> cpu out of deep C states until resolved.
..

Or perhaps some initialization on the main-boot patch
just doesn't happen on the resume-from-hibernate paths ?
(either in the BIOS or kernel or drivers ..)