ACPI; S5 state; powerbutton->clean shutdown

ACPI; S5 state; powerbutton->clean shutdown

[Wilkinson, Alex]

Hi all,

Can some please recommend to me the correct method in Linux to confirm
whether or not pressing the power-button on an ATX box triggers an ACPI
state S5 (for both 2.6 and 2.4 kernels).

In the FreeBSD world I would do something like:

   #sysctl hw.acpi.power_button_state
   hw.acpi.power_button_state: S5
   
   #grep acpi_button /var/run/dmesg.boot
   acpi_button0: <Power Button> on acpi0

Distributions are Redhat 9.0, Fedora Core 3, and SuSE Enterprise 9.2.

Cheers

 -aW


-------------------------------------------------------
The SF.Net email is sponsored by: Beat the post-holiday blues
Get a FREE limited edition SourceForge.net t-shirt from ThinkGeek.
It's fun and FREE -- well, almost....http://www.thinkgeek.com/sfshirt

Re: ACPI; S5 state; powerbutton->clean shutdown

[Matthew Garrett]

On Fri, 2005-01-07 at 11:26 +1030, Wilkinson, Alex wrote:

> Can some please recommend to me the correct method in Linux to confirm
> whether or not pressing the power-button on an ATX box triggers an ACPI
> state S5 (for both 2.6 and 2.4 kernels).

Linux has no in-kernel policy for acpi events. The userspace acpid (SuSE
have something with a different name) receives the events and can be
configured to do anything in response.

-- 
Matthew Garrett | mjg59-1xO5oi07KQx4cg9Nei1l7Q@public.gmane.org



-------------------------------------------------------
The SF.Net email is sponsored by: Beat the post-holiday blues
Get a FREE limited edition SourceForge.net t-shirt from ThinkGeek.
It's fun and FREE -- well, almost....http://www.thinkgeek.com/sfshirt

Re: ACPI; S5 state; powerbutton->clean shutdown

[Wilkinson, Alex]

    0n Fri, Jan 07, 2005 at 01:24:16AM +0000, Matthew Garrett wrote: 

    >On Fri, 2005-01-07 at 11:26 +1030, Wilkinson, Alex wrote:
    >
    >> Can some please recommend to me the correct method in Linux to confirm
    >> whether or not pressing the power-button on an ATX box triggers an ACPI
    >> state S5 (for both 2.6 and 2.4 kernels).
    >
    >Linux has no in-kernel policy for acpi events. The userspace acpid (SuSE
    >have something with a different name) receives the events and can be
    >configured to do anything in response.

man page ?
Tool to disassemble DSDT into ASL ?

 - aW


-------------------------------------------------------
The SF.Net email is sponsored by: Beat the post-holiday blues
Get a FREE limited edition SourceForge.net t-shirt from ThinkGeek.
It's fun and FREE -- well, almost....http://www.thinkgeek.com/sfshirt

Re: ACPI; S5 state; powerbutton->clean shutdown

[Matthew Garrett]

On Fri, 2005-01-07 at 12:00 +1030, Wilkinson, Alex wrote:
>     0n Fri, Jan 07, 2005 at 01:24:16AM +0000, Matthew Garrett wrote: 

>     >Linux has no in-kernel policy for acpi events. The userspace acpid (SuSE
>     >have something with a different name) receives the events and can be
>     >configured to do anything in response.
> 
> man page ?
> Tool to disassemble DSDT into ASL ?

I'm not quite sure what you're asking. You can check if the power button
will generate an ACPI event by checking whether it's present
in /proc/acpi/buttons/ - you shouldn't need to touch the DSDT. acpid
seems to have a manpage. Intel's ACPI compiler includes a disassemble
option.

-- 
Matthew Garrett | mjg59-1xO5oi07KQx4cg9Nei1l7Q@public.gmane.org



-------------------------------------------------------
The SF.Net email is sponsored by: Beat the post-holiday blues
Get a FREE limited edition SourceForge.net t-shirt from ThinkGeek.
It's fun and FREE -- well, almost....http://www.thinkgeek.com/sfshirt

Re: ACPI; S5 state; powerbutton->clean shutdown

[Wilkinson, Alex]

    0n Fri, Jan 07, 2005 at 12:46:33PM +0000, Matthew Garrett wrote: 

    >On Fri, 2005-01-07 at 12:00 +1030, Wilkinson, Alex wrote:
    >>     0n Fri, Jan 07, 2005 at 01:24:16AM +0000, Matthew Garrett wrote: 
    >
    >>     >Linux has no in-kernel policy for acpi events. The userspace acpid (SuSE
    >>     >have something with a different name) receives the events and can be
    >>     >configured to do anything in response.
    >> 
    >> man page ?
    >> Tool to disassemble DSDT into ASL ?
    >
    >I'm not quite sure what you're asking. You can check if the power button
    >will generate an ACPI event by checking whether it's present
    >in /proc/acpi/buttons/ - you shouldn't need to touch the DSDT. acpid
    >seems to have a manpage. Intel's ACPI compiler includes a disassemble
    >option.

Ok, thanks. I have a RH 9.0 (2.4.20-8smp) that does not have ACPI
support. Do 2.4 kernels support ACPI ?

If not, is it possible to configure a 2.4 system to do a clean shutdown
upon a power button  press ?

Cheers

  - aW


-------------------------------------------------------
The SF.Net email is sponsored by: Beat the post-holiday blues
Get a FREE limited edition SourceForge.net t-shirt from ThinkGeek.
It's fun and FREE -- well, almost....http://www.thinkgeek.com/sfshirt

Re: ACPI; S5 state; powerbutton->clean shutdown

[Matthew Garrett]

On Mon, 2005-01-10 at 13:40 +1030, Wilkinson, Alex wrote:

> Ok, thanks. I have a RH 9.0 (2.4.20-8smp) that does not have ACPI
> support. Do 2.4 kernels support ACPI ?

Yup. It's a compile option, though, so it's possible that it was
disabled in your kernel.

-- 
Matthew Garrett | mjg59-1xO5oi07KQx4cg9Nei1l7Q@public.gmane.org



-------------------------------------------------------
The SF.Net email is sponsored by: Beat the post-holiday blues
Get a FREE limited edition SourceForge.net t-shirt from ThinkGeek.
It's fun and FREE -- well, almost....http://www.thinkgeek.com/sfshirt

RE: ACPI; S5 state; powerbutton->clean shutdown

[Li, Shaohua]

>
>Can some please recommend to me the correct method in Linux to confirm
>whether or not pressing the power-button on an ATX box triggers an ACPI
>state S5 (for both 2.6 and 2.4 kernels).
>
Change your acpid script. In FC2, the script is in /etc/acpi/events

Thanks,
Shaohua


-------------------------------------------------------
The SF.Net email is sponsored by: Beat the post-holiday blues
Get a FREE limited edition SourceForge.net t-shirt from ThinkGeek.
It's fun and FREE -- well, almost....http://www.thinkgeek.com/sfshirt