More advanced power savings (rev. the DPMS extension).

[Jim Gettys]

On Tue, 2006-08-08 at 12:17 +0200, Egbert Eich wrote:
Jim Gettys writes:
\
>  > Because in the meanwhile, long before the screen saver has kicked in and
>  > blanked the screen, we'll have powered down the entire system (suspend
>  > to RAM) while leaving the screen on with the DCON driving it, and having
>  > to resume from RAM to just turn off the screen seemed worth the cost of
>  > a register.  So I can have the processor off long before the screen goes
>  > to a blank state, and not have to wake the processor to get the screen
>  > saver to ultimately trigger.
>  > 
>  > Moral: I don't necessarily want user space to have to handle everything.
> 
> Right. It should be a driver feature. One that will hopefully be
> configurable 'on the fly' once the infrastrucutre is there.
> 

Yes, exactly: we are building an fbdev driver right now to support this.

My point was that informing a client and having it turn off the screen,
an attitude I'd normally support, does not work well for this situation,
where we'd like to set the timeout in the hardware to turn off the
screen and then suspend the machine, avoiding a wakeup that would do
nothing more than turn off the screen.
                               - Jim


-- 
Jim Gettys
One Laptop Per Child