backlight device, suspend/resume...
Carsten Haitzler (The Rasterman)
raster at openmoko.org
Wed Jul 9 13:14:42 CEST 2008
On Tue, 08 Jul 2008 17:06:16 -0500 "Mike (mwester)" <mwester at dls.net> babbled:
> Werner Almesberger wrote:
> > Mike (mwester) wrote:
> >> In practice, compromise is necessary. Can we at least let the kernel
> >> bring the backlight up to a level that would allow the user to see vague
> >> outlines of something on the screen? Or failing that, can we make the
> >> backlight level on resume configurable in some way?
> > If you need this, why not set the backlight to "dim" instead of "off"
> > before requesting the suspend ?
> Because I don't control the source code for the user-space apps that put
> it into suspend in the first place.
> Sure, it's easy - and arguably correct - to punt on the whole problem,
> and make user-space deal with it.
imho it's best to let userspace deal with it. as it currently stands its
impossible to have userspace do anything smart as the kernel second-guesses and
thinks it knows better - no choices (backlight will be on and at full
brightness on resume, like it or not, regardless of resume reason). :)
> So what then, when the user reports that "the phone randomly powers
> off"? Do the hardware/kernel folks just tell the user to go away, and
> bother the application folks?
yup. as a user doesn't know visually the difference between a blank-screen and a
power off (until the play with the touchscreen and it doesn't come back to
life). the phone does "randomly poweroff". but its not"random" its a design.
its intended suspending when the phone is idle for a period of time.
> > The secret about successful kernel programming is to do as little as
> > possible in the kernel :-)
> The secret to customer satisfaction is proactively thinking about what
> might go wrong, and providing the user with information that might be
> helpful. Sometimes that goes in a different direction from expending
> minimal effort in the kernel.
indeed.so we should not have a splash screen. the user can't read out all the
kernel boot logs to us if something goes wrong! we need to keep them! just
being sarcastic - but you get the point. we already remove all kernel
bootmessages. i see this backlight thing as no different. :)
Carsten Haitzler (The Rasterman) <raster at openmoko.org>
More information about the openmoko-kernel