is it a timer bug?

is it a timer bug?

[Shi, Alex]

I notice the timer interrupts is quit high on the 2.6.28-rc8 kernel in my Nehalem machine. 
A, 180+/second timer interrupts in my Nehalem machine with tickless enable. 
B, the timer interrupts reduce to 10-/second with tickless and highres enable. 
On 2.6.27 kernel, the timer interrupts can keep low with tickless kernel whenever the highres feature is enable or not. 
So it is timer bug on my machine? 

Brgs
Alex 

Re: is it a timer bug?

[Arjan van de Ven]

On Mon, 15 Dec 2008 15:07:17 +0800
"Shi, Alex" <alex.shi@intel.com> wrote:

> I notice the timer interrupts is quit high on the 2.6.28-rc8 kernel
> in my Nehalem machine. A, 180+/second timer interrupts in my Nehalem
> machine with tickless enable. B, the timer interrupts reduce to
> 10-/second with tickless and highres enable. On 2.6.27 kernel, the
> timer interrupts can keep low with tickless kernel whenever the
> highres feature is enable or not. So it is timer bug on my machine? 
> 

Hi,

can you give us the powertop output of these production Core i7 
machines? That is the most likely way to get a clue about what is going
on...

Greetings,
   Arjan van de Ven

-- 
Arjan van de Ven 	Intel Open Source Technology Centre
For development, discussion and tips for power savings, 
visit http://www.lesswatts.org

RE: is it a timer bug?

[Shi, Alex]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset="gb2312", Size: 2608 bytes --]

OK, but my machine is not a production machine. 
A, 2.6.28-rc8 with tickles and highres
     PowerTOP version 1.9       (C) 2007 Intel Corporation

Cn                Avg residency       P-states (frequencies)
C0 (cpu running)        (65.1%)
C1                0.0ms ( 0.0%)
C2              105.8ms (34.9%)



Wakeups-from-idle per second :  3.8     interval: 5.0s
no ACPI power usage estimate available
B, 2.6.28-rc8 with tickless only 
     PowerTOP version 1.9       (C) 2007 Intel Corporation

Cn                Avg residency       P-states (frequencies)
C0 (cpu running)        (14.0%)
C1                0.0ms ( 0.0%)
C2                3.4ms (86.0%)



Wakeups-from-idle per second : 255.6    interval: 5.0s
no ACPI power usage estimate available

Top causes for wakeups:
  87.9% (267.6)       <interrupt> : extra timer interrupt
C, 2.6.27 with tickless only: 
    PowerTOP version 1.9       (C) 2007 Intel Corporation

Cn                Avg residency       P-states (frequencies)
C0 (cpu running)        (53.8%)         2.53 Ghz     0.0%
C1                0.0ms ( 0.0%)         2.53 Ghz    87.5%
C2              243.0ms (46.2%)         2.40 Ghz     0.0%
                                        1.60 Ghz    12.5%


Wakeups-from-idle per second :  2.2     interval: 30.0s
no ACPI power usage estimate available
I have checked the /proc/interrupts, the interrupt 0 is really triggered too much times. 
Brgs 
Alex 

>-----Original Message-----
>From: Arjan van de Ven [mailto:arjan@infradead.org]
>Sent: 2008Äê12ÔÂ15ÈÕ 16:09
>To: Shi, Alex
>Cc: linux-kernel@vger.kernel.org
>Subject: Re: is it a timer bug?
>
>On Mon, 15 Dec 2008 15:07:17 +0800
>"Shi, Alex" <alex.shi@intel.com> wrote:
>
>> I notice the timer interrupts is quit high on the 2.6.28-rc8 kernel
>> in my Nehalem machine. A, 180+/second timer interrupts in my Nehalem
>> machine with tickless enable. B, the timer interrupts reduce to
>> 10-/second with tickless and highres enable. On 2.6.27 kernel, the
>> timer interrupts can keep low with tickless kernel whenever the
>> highres feature is enable or not. So it is timer bug on my machine?
>>
>
>Hi,
>
>can you give us the powertop output of these production Core i7
>machines? That is the most likely way to get a clue about what is going
>on...
>
>Greetings,
>   Arjan van de Ven
>
>--
>Arjan van de Ven 	Intel Open Source Technology Centre
>For development, discussion and tips for power savings,
>visit http://www.lesswatts.org
ÿôèº{.nÇ+‰·Ÿ®‰­†+%ŠËÿ±éݶ\x17¥Šwÿº{.nÇ+‰·¥Š{±þG«éÿŠ{ayº\x1dʇڙë,j\a­¢f£¢·hšïêÿ‘êçz_è®\x03(­éšŽŠÝ¢j"ú\x1a¶^[m§ÿÿ¾\a«þG«éÿ¢¸?™¨è­Ú&£ø§~á¶iO•æ¬z·švØ^\x14\x04\x1a¶^[m§ÿÿÃ\fÿ¶ìÿ¢¸?–I¥

Re: is it a timer bug?

[Arjan van de Ven]

On Mon, 15 Dec 2008 16:19:37 +0800
"Shi, Alex" <alex.shi@intel.com> wrote:

> OK, but my machine is not a production machine. 
> A, 2.6.28-rc8 with tickles and highres
>      PowerTOP version 1.9       (C) 2007 Intel Corporation
> 
> Cn                Avg residency       P-states (frequencies)
> C0 (cpu running)        (65.1%)
> C1                0.0ms ( 0.0%)
> C2              105.8ms (34.9%)
> 
> 
> 
> Wakeups-from-idle per second :  3.8     interval: 5.0s
> no ACPI power usage estimate available
> B, 2.6.28-rc8 with tickless only 
>      PowerTOP version 1.9       (C) 2007 Intel Corporation
> 
> Cn                Avg residency       P-states (frequencies)
> C0 (cpu running)        (14.0%)
> C1                0.0ms ( 0.0%)
> C2                3.4ms (86.0%)
> 
> 
> 
> Wakeups-from-idle per second : 255.6    interval: 5.0s
> no ACPI power usage estimate available
> 
> Top causes for wakeups:
>   87.9% (267.6)       <interrupt> : extra timer interrupt

unfortunately you don't give more than the first line ;-(



-- 
Arjan van de Ven 	Intel Open Source Technology Centre
For development, discussion and tips for power savings, 
visit http://www.lesswatts.org