generic hotkey driver

generic hotkey driver

[Yu, Luming]

   I'm thinking if we need a generic hotkey driver that is configurable
to any specific machine. 
There already have two specific hotkey driver: asus_acpi.c and
toshiba_acpi.c. And there 
are two upcoming driver : ibm_acpi.c and panasonic_acpi.c.  I don't know
how many
hotkey driver will be invented in the future. But I know it will mess up
the code structure.
  So, I'm considering a generic hotkey driver that will export
configurable interface to 
accommodate the difference.

 a. Configure interface:
	1. register mapping of event number to internal hotkey number.
	2. register expected acpi_handle which is needed to install
notify handler, which will generate event for acpid.
	3. register the AML methods that should be executed , with
respect to  the hotkey get pressed.
 b. Expose interface:
	1. Expose the event action interface through /proc/acpi/action,
the interface will receive argument which
		will be passed into AML method, as well as AML method
should be executed.
 c. Use interface:
	1. When user space acpid get a event, it can call related AML
methods related to this hotkey.

Thanks,
Luming
	

  


-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_idP47&alloc_id\x10808&op=click

Re: generic hotkey driver

[Hiroshi Miura]

Hi,

I think hotkey drivers are  so differ that it is enough to decide
event format. ie) 'ibm KEY 00001 00001' or 'hotkey/pcc HKEY 000003 000001'

Lcd brightness control is defined in ACPI spec apendix.
I'm starting implement that.

I want to make generic lcd driver that is configurable for each extra drivers.

Bluetooth, dock, and other device settings are not common.
So these may be implement in each extra drivers. eg) ibm_acpi.c has these functions.

At Wed, 1 Sep 2004 18:32:53 +0800,
Yu, Luming wrote:
> 
>    I'm thinking if we need a generic hotkey driver that is configurable
> to any specific machine. 
> There already have two specific hotkey driver: asus_acpi.c and
> toshiba_acpi.c. And there 
> are two upcoming driver : ibm_acpi.c and panasonic_acpi.c.  I don't know
> how many
> hotkey driver will be invented in the future. But I know it will mess up
> the code structure.
>   So, I'm considering a generic hotkey driver that will export
> configurable interface to 
> accommodate the difference.
> 
>  a. Configure interface:
> 	1. register mapping of event number to internal hotkey number.
> 	2. register expected acpi_handle which is needed to install
> notify handler, which will generate event for acpid.
> 	3. register the AML methods that should be executed , with
> respect to  the hotkey get pressed.
>  b. Expose interface:
> 	1. Expose the event action interface through /proc/acpi/action,
> the interface will receive argument which
> 		will be passed into AML method, as well as AML method
> should be executed.
>  c. Use interface:
> 	1. When user space acpid get a event, it can call related AML
> methods related to this hotkey.

-- 
Hiroshi Miura  --- http://www.da-cha.org/  --- miura-yiisDzvROlQdnm+yROfE0A@public.gmane.org
NTTDATA Corp. OpenSource Software Center. --- miurahr-3MafRgGXt7BL9jVzuh4AOg@public.gmane.org 
NTTDATA Intellilink Corp. OpenSource Engineering Dev. -- miurahr-w0OK63jvRlAuJ+9fw/WgBHgSJqDPrsil@public.gmane.org
Key fingerprint = 9117 9407 5684 FBF1 4063  15B4 401D D077 04AB 8617


-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click

Re: generic hotkey driver

[John Belmonte]

Yu, Luming wrote:
>  a. Configure interface:
> 	1. register mapping of event number to internal hotkey number.
> 	2. register expected acpi_handle which is needed to install
> notify handler, which will generate event for acpid.
> 	3. register the AML methods that should be executed , with
> respect to  the hotkey get pressed.

As far as I know, there is no hotkey event available in Toshiba's case. 
  The only way to monitor the keys is by polling.

-John


-- 
http://giftfile.org/  ::  giftfile project


-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click

RE: generic hotkey driver

[Yu, Luming]

>Yu, Luming wrote:
>>  a. Configure interface:
>> 	1. register mapping of event number to internal hotkey number.
>> 	2. register expected acpi_handle which is needed to install
>> notify handler, which will generate event for acpid.
>> 	3. register the AML methods that should be executed , with
>> respect to  the hotkey get pressed.
>
>As far as I know, there is no hotkey event available in 
>Toshiba's case. 
>  The only way to monitor the keys is by polling.

Hmm. It's not efficient way for hotkey.
I'm curious what's the polling frequency.

I guess I can slightly modify the configure interface to
add a polling register interface.  To keep other parts of
interfaces unchanged, just need to generate corresponding
pseudo event for each active event found by polling.

Does it make sense?

Thanks,
Luming




-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_idP47&alloc_id\x10808&op=click

RE: generic hotkey driver

[Yu, Luming]

>
>I think hotkey drivers are  so differ that it is enough to decide
>event format. ie) 'ibm KEY 00001 00001' or 'hotkey/pcc HKEY 
>000003 000001'
>

We don't need to care about the format of event. Since it is
configurable,
we just need to understand the internal meaning. So maybe we
need to define the internal event number, which stands for a
special purpose. such as:

	001	turn off display
	002	inc brightness
	003	dec brightness
	004	s3
	......

>Lcd brightness control is defined in ACPI spec apendix.
>I'm starting implement that.
>

Be careful, there has a video extension driver 
filed at http://bugzilla.kernel.org/show_bug.cgi?id=1944

>I want to make generic lcd driver that is configurable for 
>each extra drivers.
>
>Bluetooth, dock, and other device settings are not common.
>So these may be implement in each extra drivers. eg) 
>ibm_acpi.c has these functions.
>

But , it is still meaningful to standardize the event format.

Thanks,
Luming


-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_idP47&alloc_id\x10808&op=click

Re: generic hotkey driver

[Karol Kozimor]

Thus wrote Yu, Luming:
> We don't need to care about the format of event. Since it is
> configurable,
> we just need to understand the internal meaning. So maybe we
> need to define the internal event number, which stands for a
> special purpose. such as:
> 
> 	001	turn off display
> 	002	inc brightness
> 	003	dec brightness
> 	004	s3
> 	......

Hmm, ASUS laptops support some 16 distinct events, I don't see your
solution maintainable.

> >Lcd brightness control is defined in ACPI spec apendix.
> >I'm starting implement that.
> Be careful, there has a video extension driver 
> filed at http://bugzilla.kernel.org/show_bug.cgi?id=1944

Sadly, most OEMs don't give a damn about the spec and implement this in
their own way.

> >Bluetooth, dock, and other device settings are not common.
> >So these may be implement in each extra drivers. eg) 
> >ibm_acpi.c has these functions.
> But , it is still meaningful to standardize the event format.

Sure, but what can be done apart from agreeing on the name of the first
field?
Best regards,

-- 
Karol 'sziwan' Kozimor
sziwan-DETuoxkZsSqrDJvtcaxF/A@public.gmane.org


-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click

Re: generic hotkey driver

[John Belmonte]

Yu, Luming wrote:
>>> a. Configure interface:
>>>	1. register mapping of event number to internal hotkey number.
>>>	2. register expected acpi_handle which is needed to install
>>>notify handler, which will generate event for acpid.
>>>	3. register the AML methods that should be executed , with
>>>respect to  the hotkey get pressed.
>>
>>As far as I know, there is no hotkey event available in 
>>Toshiba's case. 
>> The only way to monitor the keys is by polling.
> 
> Hmm. It's not efficient way for hotkey.
> I'm curious what's the polling frequency.

I don't poll within the kernel-- I leave that issue to userspace.  The 
hotkeys are queued, and I've found a period of .2s to be more than adequate.

> I guess I can slightly modify the configure interface to
> add a polling register interface.  To keep other parts of
> interfaces unchanged, just need to generate corresponding
> pseudo event for each active event found by polling.
> 
> Does it make sense?

If you want to provide hotkey events to userspace, I guess there is no 
other choice.

-John


-- 
http://giftfile.org/  ::  giftfile project


-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click

RE: generic hotkey driver

[Yu, Luming]

 
>> We don't need to care about the format of event. Since it is
>> configurable,
>> we just need to understand the internal meaning. So maybe we
>> need to define the internal event number, which stands for a
>> special purpose. such as:
>> 
>> 	001	turn off display
>> 	002	inc brightness
>> 	003	dec brightness
>> 	004	s3
>> 	......
>
>Hmm, ASUS laptops support some 16 distinct events, I don't see your
>solution maintainable.
>

Hmm, I will maintain a table internally, which will keep the mapping
between
the raw event value with the standard internal event num. If the raw
event value
cannot be mapped to any known standard event num, then the user can
treat 
it as extended event. The entry could be:

struct raw_value_2_event {
	int	raw_event_number;
	int	standard_event;
	int	raw_event_flags;
	ACTION_LIST	*AML_METHOD;
}

The user or init script can input the mapping against the actual laptop
through
the interface:
/proc/acpi/hotkey/config/raw_event_mapping

To register expected acpi_handle which is needed to install
notify handler, which will generate event through /proc/acpi/event for
acpid.
we can input AML pathname on which driver will generate event
when get notified in AML method, like Notify( VGA, 0x80), through 

/proc/acpi/hotkey/config/bus_notify

To register polling methods which will generate pseudo event, we can
input the expected AML methods pathname through
/proc/acpi/hotkey/config/pulling_method

Each successful pull will generate a event through /proc/acpi/event to
notify user land to get pulling data from /proc/acpi/hotkey/pulling_data
And the userland daemon will interpret the actual meaning of the raw
data.

To register the AML methods that should be executed , with
respect to  the hotkey get pressed. we can input AML pathname through
/proc/acpi/hotkey/config/event_action.

When userland daemon received a event, the daemon just need to feed
/proc/acpi/hotkey/action_argument with the proper argument string.
A "" means no argument is need for the method.

This is just a rough idea. 

Thanks,
Luming


-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_idP47&alloc_id\x10808&op=click

Re: generic hotkey driver

[Pavel Machek]

Hi!

> >I think hotkey drivers are  so differ that it is enough to decide
> >event format. ie) 'ibm KEY 00001 00001' or 'hotkey/pcc HKEY 
> >000003 000001'
> >
> 
> We don't need to care about the format of event. Since it is
> configurable,
> we just need to understand the internal meaning. So maybe we
> need to define the internal event number, which stands for a
> special purpose. such as:
> 
> 	001	turn off display
> 	002	inc brightness
> 	003	dec brightness
> 	004	s3
> 	......
> 

If those are keys, you should just treat them as keys... That is hook
it into input subsystem.
								Pavel
-- 
People were complaining that M$ turns users into beta-testers...
...jr ghea gurz vagb qrirybcref, naq gurl frrz gb yvxr vg gung jnl!


-------------------------------------------------------
This SF.Net email is sponsored by: thawte's Crypto Challenge Vl
Crack the code and win a Sony DCRHC40 MiniDV Digital Handycam
Camcorder. More prizes in the weekly Lunch Hour Challenge.
Sign up NOW http://ad.doubleclick.net/clk;10740251;10262165;m