Trying to generate ACPI events from software

Trying to generate ACPI events from software

[Angela Dalton]

Hello,

I apologize in advance for the lengthy message!

I am working on a program in which I would like to control the video 
device power state based on something other than a timeout (specifically 
based on information from external sensors).  Unfortunately, I can't 
seem to figure out how to turn off the LCD and backlight from the 
program.  I can blank the screen using "xset dpms force off" but the 
backlight remains on, and a subsequent "xset dpms force on" from my 
program has no effect, I actually have to physically move the mouse or 
press a key.  I also tried writing to the /dev/nvram locations that 
change when the lid is closed or the special keys are pressed to switch 
from LCD to CRT.  Apparently, the changes in the nvram values do not 
directly cause the display state change.

I have been looking through the ACPI driver source code and the ACPID 
code as well as the APCI Specification.  I have not been able to figure 
out how to generate an event from within my program to tell ACPI to shut 
the LCD off on my ThinkPad.  Here is what I think I understand:

1. Based on the ACPI Specification, there is a Notify(VGA, 0x80) call to 
tell the OSPM when the user presses the hotkey (in my case, FnF7) for 
switching monitor output.  The OSPM should call the _DSS method to set 
the device state of the LCD and CRT devices.

2. From looking through the driver code, I found a file evevent.c that 
seems to handle General Purpose events such as the one I want to cause. 
 This code detects events like a power button event and dispatches it to 
an installed handler.  There appear to be both fixed event handlers and 
generic general purpose event handlers.  There are also notify handlers. 
 This is a confusing part of the code for me as far as what I need to 
do.  Do I need to write a notify handler and a fixed event handler?  Is 
there a general notify handler that will invoke a general purpose event 
handler that I would write?

3.  Finally, based on what I see in the ospm directory, there is code to 
handle a variety of events.  I assume this would be where my handler 
would go.  It seems like this is a rather large undertaking, based on 
the code for the other devices.  

My questions are:

Am I on the right track or completely lost??  And is there is much 
easier way to do this using the ACPID or OSPMD?  If so, can someone 
please explain it to me?  One thought I had was if my program could 
generate a lid event then it might shut off the display (but how would I 
get it back on and how can I prevent other stuff from also occuring, 
like a suspend or hibernate?)

I appreciate any help I can get!

Best Regards,
Angela Dalton



-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf

RE: Trying to generate ACPI events from software

[Yu, Luming]

Do you want ACPI export a interface to user to shut off the LCD through command like:
#echo -n 3  > /proc/acpi/display/LCD0
I don't think there is such functionality in current ACPI implementation.
Maybe you can try to write such a driver display. 
But we should enhance the stability of ACPI core on various platform first.
Anyway, the various attractive PM features are based on a stable ACPI core.

Thanks,
Luming



-----Original Message-----
From: Angela Dalton [mailto:angela-VPOBD8XdS/WVc3sceRu5cw@public.gmane.org]
Sent: 2003?4?17? 21:36
To: acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
Subject: [ACPI] Trying to generate ACPI events from software


Hello,

I apologize in advance for the lengthy message!

I am working on a program in which I would like to control the video 
device power state based on something other than a timeout (specifically 
based on information from external sensors).  Unfortunately, I can't 
seem to figure out how to turn off the LCD and backlight from the 
program.  I can blank the screen using "xset dpms force off" but the 
backlight remains on, and a subsequent "xset dpms force on" from my 
program has no effect, I actually have to physically move the mouse or 
press a key.  I also tried writing to the /dev/nvram locations that 
change when the lid is closed or the special keys are pressed to switch 
from LCD to CRT.  Apparently, the changes in the nvram values do not 
directly cause the display state change.

I have been looking through the ACPI driver source code and the ACPID 
code as well as the APCI Specification.  I have not been able to figure 
out how to generate an event from within my program to tell ACPI to shut 
the LCD off on my ThinkPad.  Here is what I think I understand:

1. Based on the ACPI Specification, there is a Notify(VGA, 0x80) call to 
tell the OSPM when the user presses the hotkey (in my case, FnF7) for 
switching monitor output.  The OSPM should call the _DSS method to set 
the device state of the LCD and CRT devices.

2. From looking through the driver code, I found a file evevent.c that 
seems to handle General Purpose events such as the one I want to cause. 
 This code detects events like a power button event and dispatches it to 
an installed handler.  There appear to be both fixed event handlers and 
generic general purpose event handlers.  There are also notify handlers. 
 This is a confusing part of the code for me as far as what I need to 
do.  Do I need to write a notify handler and a fixed event handler?  Is 
there a general notify handler that will invoke a general purpose event 
handler that I would write?

3.  Finally, based on what I see in the ospm directory, there is code to 
handle a variety of events.  I assume this would be where my handler 
would go.  It seems like this is a rather large undertaking, based on 
the code for the other devices.  

My questions are:

Am I on the right track or completely lost??  And is there is much 
easier way to do this using the ACPID or OSPMD?  If so, can someone 
please explain it to me?  One thought I had was if my program could 
generate a lid event then it might shut off the display (but how would I 
get it back on and how can I prevent other stuff from also occuring, 
like a suspend or hibernate?)

I appreciate any help I can get!

Best Regards,
Angela Dalton



-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
Acpi-devel mailing list
Acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
https://lists.sourceforge.net/lists/listinfo/acpi-devel


-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf

Re: Trying to generate ACPI events from software

[Angela Dalton]

This is for a special project and will not be a normal part of ACPI, but 
I need the ability to turn off the display
for my project to work.  So I guess I will need to write a handler for 
the video devices in the acpi driver?

Yu, Luming wrote:

>Do you want ACPI export a interface to user to shut off the LCD through command like:
>#echo -n 3  > /proc/acpi/display/LCD0
>I don't think there is such functionality in current ACPI implementation.
>Maybe you can try to write such a driver display. 
>But we should enhance the stability of ACPI core on various platform first.
>Anyway, the various attractive PM features are based on a stable ACPI core.
>
>Thanks,
>Luming
>
>
>
>-----Original Message-----
>From: Angela Dalton [mailto:angela-VPOBD8XdS/WVc3sceRu5cw@public.gmane.org]
>Sent: 2003?4?17? 21:36
>To: acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
>Subject: [ACPI] Trying to generate ACPI events from software
>
>
>Hello,
>
>I apologize in advance for the lengthy message!
>
>I am working on a program in which I would like to control the video 
>device power state based on something other than a timeout (specifically 
>based on information from external sensors).  Unfortunately, I can't 
>seem to figure out how to turn off the LCD and backlight from the 
>program.  I can blank the screen using "xset dpms force off" but the 
>backlight remains on, and a subsequent "xset dpms force on" from my 
>program has no effect, I actually have to physically move the mouse or 
>press a key.  I also tried writing to the /dev/nvram locations that 
>change when the lid is closed or the special keys are pressed to switch 
>from LCD to CRT.  Apparently, the changes in the nvram values do not 
>directly cause the display state change.
>
>I have been looking through the ACPI driver source code and the ACPID 
>code as well as the APCI Specification.  I have not been able to figure 
>out how to generate an event from within my program to tell ACPI to shut 
>the LCD off on my ThinkPad.  Here is what I think I understand:
>
>1. Based on the ACPI Specification, there is a Notify(VGA, 0x80) call to 
>tell the OSPM when the user presses the hotkey (in my case, FnF7) for 
>switching monitor output.  The OSPM should call the _DSS method to set 
>the device state of the LCD and CRT devices.
>
>2. From looking through the driver code, I found a file evevent.c that 
>seems to handle General Purpose events such as the one I want to cause. 
> This code detects events like a power button event and dispatches it to 
>an installed handler.  There appear to be both fixed event handlers and 
>generic general purpose event handlers.  There are also notify handlers. 
> This is a confusing part of the code for me as far as what I need to 
>do.  Do I need to write a notify handler and a fixed event handler?  Is 
>there a general notify handler that will invoke a general purpose event 
>handler that I would write?
>
>3.  Finally, based on what I see in the ospm directory, there is code to 
>handle a variety of events.  I assume this would be where my handler 
>would go.  It seems like this is a rather large undertaking, based on 
>the code for the other devices.  
>
>My questions are:
>
>Am I on the right track or completely lost??  And is there is much 
>easier way to do this using the ACPID or OSPMD?  If so, can someone 
>please explain it to me?  One thought I had was if my program could 
>generate a lid event then it might shut off the display (but how would I 
>get it back on and how can I prevent other stuff from also occuring, 
>like a suspend or hibernate?)
>
>I appreciate any help I can get!
>
>Best Regards,
>Angela Dalton
>
>
>
>-------------------------------------------------------
>This sf.net email is sponsored by:ThinkGeek
>Welcome to geek heaven.
>http://thinkgeek.com/sf
>_______________________________________________
>Acpi-devel mailing list
>Acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
>https://lists.sourceforge.net/lists/listinfo/acpi-devel
>  
>




-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf