preparing pcf50633 for upstream

Mike (mwester) mwester at dls.net
Wed Oct 1 21:50:06 CEST 2008


Werner Almesberger wrote:
...
> So, two questions:
> 1) is APM emulation something new code should use at all ?

I can't say I particularly like the APM interface.  I'm unclear on the
alternative -- is there another standard, something that works exactly
the same on the GTA01, as well as with a non-coulomb-counting battery in
the GTA02?  Or does something need to be invented?  Or does user-space
need to figure it out for each device and battery type?

> 2) would user space developers cry bitter tears if we removed it ?

Dunno.  Sadly, user-space has splintered into several different images,
with variants on those (someone mentioned eight distinct images the
other day) so the answer might be different for different folks.  But I
tend to think that if a single, uniform alternative to APM exists, there
would be few tears; APM is very limited.

If we can't have a single uniform interface, then I think we can
alleviate the pain and possible inconsistencies in reporting battery
state if we can publish a recommended algorithm (e.g. what sysfs files
to read, or what kernel events to listen for, on each device and battery).

And if none of the above makes any sense, I've just proven how confusing
this all is to me!


Mike (mwester)



More information about the openmoko-kernel mailing list