Re: 2 day GENERIC-current eat 2 CPU core at 100%

Andrey Smagin wrote:
> Hi, yesterday I tried switch  event timer on i8254 - it do nothing.
> I disabled hyperthreading - now eat from 50% to 100%
> All dmesg is lines:
> (noperiph:ata3:0:-1:-1): rescan already queued
> (noperiph:ata3:0:-1:-1): rescan already queued
> (noperiph:ata2:0:-1:-1): rescan already queued
> (noperiph:ata3:0:-1:-1): rescan already queued
> (noperiph:ata3:0:-1:-1): rescan already queued
> ....
> I  boot FreeBSD from USB with no ATA - may be it is.

These messages are not what I expected to see, but they tell me that
your problem may be SATA related. I've got the same Intel D525MW board
and think reproduced the problem. I hope I've even fixed it. :) Retry
please with fresh CURRENT sources or at least with this patch applied:
http://svn.freebsd.org/changeset/base/222897

> Wed, 08 Jun 2011 20:34:42 +0300 письмо от Alexander Motin <mav_at_FreeBSD.org>:
>> On 07.06.2011 20:12, Andrey Smagin wrote:
>>> vmstat -i
>>> interrupt total rate
>>> irq16: uhci3   205  0
>>> irq20: hpet0 147924380  1126
>>> irq23: uhci0 ehci0   522517     3
>>> total                    148447102   1130
>>>
>>> Tue, 7 Jun 2011 10:34:01 +0200 письмо от Hans Petter
>> Selasky<hselasky_at_c2i.net>:
>>>> On Tuesday 07 June 2011 10:09:47 Andrey Smagin wrote:
>>>>> I upgraded 2 day ago from 2010xxxx-current box on Intel D525MW.
>>>>> System very slow down after that.
>>>>> kern.hz=50
>>>>> in systat -vmstat - 140hpet interrupts/s
>>>>> at top 25% in interrupts 25% in system
>>>>> because hyperthreading system found 4 cpu.
>>>> What does vmstat -i output?
>> Send me please full verbose dmesg and output of the `sysctl 
>> kern.eventtimer` and `sysctl kern.timecounter`.
>>
>> Try to switch to another timer:
>> sysctl kern.eventtimer.timer=LAPIC
>>
>> Try to switch to periodic timers (instead):
>> sysctl kern.eventtimer.periodic=1

-- 
Alexander Motin