acpi_power_off doesn't

acpi_power_off doesn't

[Harald Dunkel]

[-- Attachment #1: Type: text/plain, Size: 594 bytes --]

Hi folks,

Sometimes acpi_power_off doesn't switch off my PC. If I
boot and press the power button again as soon as it is
up, then it shuts down as expected. But if I login, let
it run for some time, logout, and press the power button,
then it walks through the runlevel 0 scripts as before,
finally it says

	acpi_power_off called.

and keeps on running. The PC does _not_ go off.

Kernel is 2.6.16.5, Pentium M 2GHz, Aopen MZ915-M.

Any idea?

This PC is supposed to startup and shutdown without
watching, so any help would be highly appreciated.


Regards

Harri



[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 252 bytes --]

acpi_power_off doesn't

[Harald Dunkel]

[-- Attachment #1: Type: text/plain, Size: 303 bytes --]

Hi folks,

Sometimes when I run 'halt' my PC does not go off. Last
words are

	acpi_power_off called

But the PC stays on.

What is the story here? I've seen this problem come up
several times, but without solution, as it seems. Any
hint would be very helpfull.


Many thanx

Harri



[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 252 bytes --]

RE: acpi_power_off doesn't

[Brown, Len]

>Sometimes when I run 'halt' my PC does not go off. Last
>words are
>
>	acpi_power_off called
>
>But the PC stays on.
>
>What is the story here? I've seen this problem come up
>several times, but without solution, as it seems. Any
>hint would be very helpfull.

Does this happen all the time, or just some of the time?
Has this always failed on box X, or did it used to
work in some release Y, and broke in some release Z?

Please supply X, Y, Z.

thanks,
-Len

RE: acpi_power_off doesn't

[Brown, Len]

>Sometimes when I run 'halt' my PC does not go off. Last
>words are
>
>	acpi_power_off called
>
>But the PC stays on.
>
>What is the story here? I've seen this problem come up
>several times, but without solution, as it seems. Any
>hint would be very helpfull.

Does this happen all the time, or just some of the time?
Has this always failed on box X, or did it used to
work in some release Y, and broke in some release Z?

Please supply X, Y, Z.

thanks,
-Len

Re: acpi_power_off doesn't

[Harald Dunkel]

[-- Attachment #1: Type: text/plain, Size: 1140 bytes --]

Hi Len,

The problem does not exist, if I boot my PC and then
halt it immediately. If I login and use it for some
time, then acpi_power_off does not work.

Box 'X' is an Aopen MZ-915M, CPU is a 2 GHz Pentium
M. It is running Debian Sid, kernel is vanilla

Linux bugs 2.6.17-rc4 #1 PREEMPT Sat May 13 16:22:54 CEST 2006 i686 GNU/Linux

Old kernels don't work on this PC due to missing
hardware support. The first vanilla kernel that worked
reliably on this box (except for acpi_power_off) was 2.6.16.


Regards

Harri
============================================================
Brown, Len wrote:
>> Sometimes when I run 'halt' my PC does not go off. Last
>> words are
>>
>> 	acpi_power_off called
>>
>> But the PC stays on.
>>
>> What is the story here? I've seen this problem come up
>> several times, but without solution, as it seems. Any
>> hint would be very helpfull.
> 
> Does this happen all the time, or just some of the time?
> Has this always failed on box X, or did it used to
> work in some release Y, and broke in some release Z?
> 
> Please supply X, Y, Z.
> 
> thanks,
> -Len
> 



[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 252 bytes --]

Re: acpi_power_off doesn't

[Robert Hancock]

Harald Dunkel wrote:
> Hi Len,
> 
> The problem does not exist, if I boot my PC and then
> halt it immediately. If I login and use it for some
> time, then acpi_power_off does not work.
> 
> Box 'X' is an Aopen MZ-915M, CPU is a 2 GHz Pentium
> M. It is running Debian Sid, kernel is vanilla
> 
> Linux bugs 2.6.17-rc4 #1 PREEMPT Sat May 13 16:22:54 CEST 2006 i686 GNU/Linux
> 
> Old kernels don't work on this PC due to missing
> hardware support. The first vanilla kernel that worked
> reliably on this box (except for acpi_power_off) was 2.6.16.

Do you get any ACPI execution errors, etc. in the dmesg output after the 
system has been running for a while? I've seen this happen after the 
ACPI machinery gets into a bad state..

-- 
Robert Hancock      Saskatoon, SK, Canada
To email, remove "nospam" from hancockr@nospamshaw.ca
Home Page: http://www.roberthancock.com/

RE: acpi_power_off doesn't

[Brown, Len]

>>> Sometimes when I run 'halt' my PC does not go off. Last
>>> words are
>>>
>>> 	acpi_power_off called
>>>
>>> But the PC stays on.
>>>
>>> What is the story here? I've seen this problem come up
>>> several times, but without solution, as it seems. Any
>>> hint would be very helpfull.
>> 
>> Does this happen all the time, or just some of the time?
>> Has this always failed on box X, or did it used to
>> work in some release Y, and broke in some release Z?
>> 
>> Please supply X, Y, Z.


>The problem does not exist, if I boot my PC and then
>halt it immediately. If I login and use it for some
>time, then acpi_power_off does not work.
>
>Box 'X' is an Aopen MZ-915M, CPU is a 2 GHz Pentium
>M. It is running Debian Sid, kernel is vanilla
>
>Linux bugs 2.6.17-rc4 #1 PREEMPT Sat May 13 16:22:54 CEST 2006 
>i686 GNU/Linux
>
>Old kernels don't work on this PC due to missing
>hardware support. The first vanilla kernel that worked
>reliably on this box (except for acpi_power_off) was 2.6.16.

If you append a "3" to the cmdline and boot without a GUI,
does the system still halt properly (even after you log in
and use it for a while)?  There may be some interaction
between X and the video hardware and system shutdown.

-Len

RE: acpi_power_off doesn't

[Brown, Len]

>>> Sometimes when I run 'halt' my PC does not go off. Last
>>> words are
>>>
>>> 	acpi_power_off called
>>>
>>> But the PC stays on.
>>>
>>> What is the story here? I've seen this problem come up
>>> several times, but without solution, as it seems. Any
>>> hint would be very helpfull.
>> 
>> Does this happen all the time, or just some of the time?
>> Has this always failed on box X, or did it used to
>> work in some release Y, and broke in some release Z?
>> 
>> Please supply X, Y, Z.


>The problem does not exist, if I boot my PC and then
>halt it immediately. If I login and use it for some
>time, then acpi_power_off does not work.
>
>Box 'X' is an Aopen MZ-915M, CPU is a 2 GHz Pentium
>M. It is running Debian Sid, kernel is vanilla
>
>Linux bugs 2.6.17-rc4 #1 PREEMPT Sat May 13 16:22:54 CEST 2006 
>i686 GNU/Linux
>
>Old kernels don't work on this PC due to missing
>hardware support. The first vanilla kernel that worked
>reliably on this box (except for acpi_power_off) was 2.6.16.

If you append a "3" to the cmdline and boot without a GUI,
does the system still halt properly (even after you log in
and use it for a while)?  There may be some interaction
between X and the video hardware and system shutdown.

-Len

Re: acpi_power_off doesn't

[Harald Dunkel]

[-- Attachment #1: Type: text/plain, Size: 2454 bytes --]

Hi Robert,

Robert Hancock wrote:
> 
> Do you get any ACPI execution errors, etc. in the dmesg output after the
> system has been running for a while? I've seen this happen after the
> ACPI machinery gets into a bad state..
> 
Nothing suspicious, as it seems. If I grep -i acpi in syslog, then I get:

May 15 13:09:11 bugs kernel: ACPI: PCI Interrupt 0000:00:1b.0[A] -> Link [LNKA] -> GSI 10 (level, low) -> IRQ 10
May 15 13:09:11 bugs kernel: ACPI: Power Button (FF) [PWRF]
May 15 13:09:11 bugs kernel: ACPI: Power Button (CM) [PWRB]
May 15 13:13:11 bugs kernel: ACPI: PCI Interrupt 0000:00:1b.0[A] -> Link [LNKA] -> GSI 10 (level, low) -> IRQ 10
May 15 13:13:11 bugs kernel: ACPI: Power Button (FF) [PWRF]
May 15 13:13:11 bugs kernel: ACPI: Power Button (CM) [PWRB]
May 15 13:18:52 bugs kernel: ACPI: Power Button (FF) [PWRF]
May 15 13:18:52 bugs kernel: ACPI: Power Button (CM) [PWRB]
May 15 13:22:46 bugs kernel: ACPI: PCI Interrupt 0000:00:1b.0[A] -> Link [LNKA] -> GSI 10 (level, low) -> IRQ 10
May 15 13:22:46 bugs kernel: ACPI: Power Button (FF) [PWRF]
May 15 13:22:46 bugs kernel: ACPI: Power Button (CM) [PWRB]
May 15 13:26:40 bugs kernel: ACPI: PCI Interrupt 0000:00:1b.0[A] -> Link [LNKA] -> GSI 10 (level, low) -> IRQ 10
May 15 13:26:40 bugs kernel: ACPI: Power Button (FF) [PWRF]
May 15 13:26:40 bugs kernel: ACPI: Power Button (CM) [PWRB]
May 15 13:51:58 bugs kernel: ACPI: PCI Interrupt 0000:01:00.0[A] -> Link [LNKA] -> GSI 10 (level, low) -> IRQ 10
May 15 14:00:50 bugs kernel: ACPI: PCI Interrupt 0000:00:1b.0[A] -> Link [LNKA] -> GSI 10 (level, low) -> IRQ 10
May 15 14:00:50 bugs kernel: ACPI: Power Button (FF) [PWRF]
May 15 14:00:50 bugs kernel: ACPI: Power Button (CM) [PWRB]
May 15 14:01:24 bugs kernel: ACPI: PCI Interrupt 0000:01:00.0[A] -> Link [LNKA] -> GSI 10 (level, low) -> IRQ 10
May 15 19:08:02 bugs kernel: ACPI: PCI Interrupt 0000:00:1b.0[A] -> Link [LNKA] -> GSI 10 (level, low) -> IRQ 10
May 15 19:08:02 bugs kernel: ACPI: Power Button (FF) [PWRF]
May 15 19:08:02 bugs kernel: ACPI: Power Button (CM) [PWRB]
May 15 19:16:45 bugs kernel: ACPI: PCI Interrupt 0000:00:1b.0[A] -> Link [LNKA] -> GSI 10 (level, low) -> IRQ 10
May 15 19:16:46 bugs kernel: ACPI: Power Button (FF) [PWRF]
May 15 19:16:46 bugs kernel: ACPI: Power Button (CM) [PWRB]
May 15 20:36:35 bugs kernel: ACPI: PCI Interrupt 0000:01:00.0[A] -> Link [LNKA] -> GSI 10 (level, low) -> IRQ 10


Regards

Harri



[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 252 bytes --]

RE: acpi_power_off doesn't

[Byron Hillis]

Brown, Len wrote:

>>>> Sometimes when I run 'halt' my PC does not go off. Last
>>>> words are
>>>>
>>>> acpi_power_off called
>>>>
>>>> But the PC stays on.
>>>>
>>>> What is the story here? I've seen this problem come up
>>>> several times, but without solution, as it seems. Any
>>>> hint would be very helpfull.
>>> 

I can second that this happens here as well. I've tried removing all
modules that I can think of and for some reason, it sometimes won't
shutdown. Strangely though, when I use suspend2 patches and hibernate
the machine, it successfully powers down when I hibernate, but still 
won't shutdown regularly using shutdown -h now.

This is with a 2.6.16 kernel on a laptop with an Intel ICH6 board.

Byron