linux-acpi.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Paweł Staszewski" <pstaszewski@itcare.pl>
To: Len Brown <lenb@kernel.org>
Cc: linux-acpi@vger.kernel.org
Subject: Re: Strange CPU load - acpi_os_read_port
Date: Thu, 19 Aug 2010 11:11:50 +0200	[thread overview]
Message-ID: <4C6CF556.7030301@itcare.pl> (raw)
In-Reply-To: <alpine.LFD.2.00.1008150958320.2313@localhost.localdomain>

W dniu 2010-08-15 16:05, Len Brown pisze:
>>>>                2217.00 13.7% acpi_os_read_port
>>>>
>>>>          
>>> curious.
>>> does /proc/interrupts show you are receiving acpi interrupts?
>>> if yes, what do you see with 'grep . /sys/firmware/acpi/interrupts/*'?
>>>
>>>        
cat /proc/interrupts | grep acpi
    9:          0          0          0          0          0          
0          0          0          0          0          0          
0          0          0          0          0   IO-APIC-fasteoi   acpi

No interrupts

/sys/firmware/acpi/interrupts/error:       0
/sys/firmware/acpi/interrupts/ff_gbl_lock:       0      enabled
/sys/firmware/acpi/interrupts/ff_pmtimer:       0       invalid
/sys/firmware/acpi/interrupts/ff_pwr_btn:       0       enabled
/sys/firmware/acpi/interrupts/ff_rt_clk:       0        disabled
/sys/firmware/acpi/interrupts/ff_slp_btn:       0       invalid
/sys/firmware/acpi/interrupts/gpe00:       0    invalid
/sys/firmware/acpi/interrupts/gpe01:       0    invalid
/sys/firmware/acpi/interrupts/gpe02:       0    disabled
/sys/firmware/acpi/interrupts/gpe03:       0    disabled
/sys/firmware/acpi/interrupts/gpe04:       0    disabled
/sys/firmware/acpi/interrupts/gpe05:       0    disabled
/sys/firmware/acpi/interrupts/gpe06:       0    enabled
/sys/firmware/acpi/interrupts/gpe07:       0    invalid
/sys/firmware/acpi/interrupts/gpe08:       0    enabled
/sys/firmware/acpi/interrupts/gpe09:       0    disabled
/sys/firmware/acpi/interrupts/gpe0A:       0    enabled
/sys/firmware/acpi/interrupts/gpe0B:       0    disabled
/sys/firmware/acpi/interrupts/gpe0C:       0    disabled
/sys/firmware/acpi/interrupts/gpe0D:       0    disabled
/sys/firmware/acpi/interrupts/gpe0E:       0    disabled
/sys/firmware/acpi/interrupts/gpe0F:       0    invalid
/sys/firmware/acpi/interrupts/gpe10:       0    invalid
/sys/firmware/acpi/interrupts/gpe11:       0    invalid
/sys/firmware/acpi/interrupts/gpe12:       0    invalid
/sys/firmware/acpi/interrupts/gpe13:       0    invalid
/sys/firmware/acpi/interrupts/gpe14:       0    invalid
/sys/firmware/acpi/interrupts/gpe15:       0    invalid
/sys/firmware/acpi/interrupts/gpe16:       0    invalid
/sys/firmware/acpi/interrupts/gpe17:       0    invalid
/sys/firmware/acpi/interrupts/gpe18:       0    invalid
/sys/firmware/acpi/interrupts/gpe19:       0    invalid
/sys/firmware/acpi/interrupts/gpe1A:       0    invalid
/sys/firmware/acpi/interrupts/gpe1B:       0    invalid
/sys/firmware/acpi/interrupts/gpe1C:       0    invalid
/sys/firmware/acpi/interrupts/gpe1D:       0    invalid
/sys/firmware/acpi/interrupts/gpe1E:       0    enabled
/sys/firmware/acpi/interrupts/gpe1F:       0    invalid
/sys/firmware/acpi/interrupts/gpe20:       0    disabled
/sys/firmware/acpi/interrupts/gpe21:       0    invalid
/sys/firmware/acpi/interrupts/gpe22:       0    invalid
/sys/firmware/acpi/interrupts/gpe23:       0    invalid
/sys/firmware/acpi/interrupts/gpe24:       0    invalid
/sys/firmware/acpi/interrupts/gpe25:       0    invalid
/sys/firmware/acpi/interrupts/gpe26:       0    invalid
/sys/firmware/acpi/interrupts/gpe27:       0    invalid
/sys/firmware/acpi/interrupts/gpe28:       0    invalid
/sys/firmware/acpi/interrupts/gpe29:       0    invalid
/sys/firmware/acpi/interrupts/gpe2A:       0    invalid
/sys/firmware/acpi/interrupts/gpe2B:       0    invalid
/sys/firmware/acpi/interrupts/gpe2C:       0    invalid
/sys/firmware/acpi/interrupts/gpe2D:       0    invalid
/sys/firmware/acpi/interrupts/gpe2E:       0    invalid
/sys/firmware/acpi/interrupts/gpe2F:       0    invalid
/sys/firmware/acpi/interrupts/gpe30:       0    invalid
/sys/firmware/acpi/interrupts/gpe31:       0    invalid
/sys/firmware/acpi/interrupts/gpe32:       0    invalid
/sys/firmware/acpi/interrupts/gpe33:       0    invalid
/sys/firmware/acpi/interrupts/gpe34:       0    invalid
/sys/firmware/acpi/interrupts/gpe35:       0    invalid
/sys/firmware/acpi/interrupts/gpe36:       0    invalid
/sys/firmware/acpi/interrupts/gpe37:       0    invalid
/sys/firmware/acpi/interrupts/gpe38:       0    invalid
/sys/firmware/acpi/interrupts/gpe39:       0    invalid
/sys/firmware/acpi/interrupts/gpe3A:       0    invalid
/sys/firmware/acpi/interrupts/gpe3B:       0    invalid
/sys/firmware/acpi/interrupts/gpe3C:       0    invalid
/sys/firmware/acpi/interrupts/gpe3D:       0    invalid
/sys/firmware/acpi/interrupts/gpe3E:       0    invalid
/sys/firmware/acpi/interrupts/gpe3F:       0    invalid
/sys/firmware/acpi/interrupts/gpe_all:       0
/sys/firmware/acpi/interrupts/sci:       0
/sys/firmware/acpi/interrupts/sci_not:       0

>>> what do you see with
>>> grep . /sys/devices/system/cpu/cpu*/cpuidle/*/*
>>>
>>>        
ls /sys/devices/system/cpu/cpu0/
cache  crash_notes  node0  thermal_throttle  topology

no cpuidle dir.
grep . /sys/devices/system/cpu/cpuidle/*
/sys/devices/system/cpu/cpuidle/current_driver:acpi_idle
/sys/devices/system/cpu/cpuidle/current_governor_ro:ladder

>>> and if there are IO states, any change if you boot with "idle=halt" or
>>> "idle=mwait"?
>>>        
>    
>> I can try to reproduce this again but it will take some time.
>>
>> For now the solution for me was:
>>
>> Change configuration for this host and boot with processor.max_cstate=0
>>      
> likely it will come back when you remove processor.max_cstate=0,
> and it will go away if you use either idle=halt or idle=mwait or
> processor.max_cstate=1, which are effectively the same.
> (and all preferable to processor.max_cstate-0 from a power
> saving point of view)
>
> If this is true, then you should try processor.max_cstate=N
> where N is 2, 3 etc, depending on the states shown in the
> cpuidle grep above.
>
> If you open a bugzilla and attach the output from acpidump,
> that may also be helpful.  If this is due to c-states, the question
> is if for some reason we are not using them optimally.
> If we already are doing what the platform allows, the
> route might be to use PM_QOS to disable states that
> are too expensive.
> (Documentation/power/pm_qos_interface.txt)
>
> cheers,
> Len Brown, Intel Open Source Technology Center.
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>
>
>    


      reply	other threads:[~2010-08-19  9:18 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-07-22 19:39 Strange CPU load - acpi_os_read_port Paweł Staszewski
2010-07-22 19:44 ` Paweł Staszewski
2010-08-15  3:33 ` Len Brown
2010-08-15 10:25   ` Paweł Staszewski
2010-08-15 14:05     ` Len Brown
2010-08-19  9:11       ` Paweł Staszewski [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=4C6CF556.7030301@itcare.pl \
    --to=pstaszewski@itcare.pl \
    --cc=lenb@kernel.org \
    --cc=linux-acpi@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).