[PATCH 11/13] [ARM] pxa/balloon3: PCF857x GPIO expander and LEDs

[marek.vasut@gmail.com (Marek Vasut)]

Dne P? 30. ?ervence 2010 15:07:55 Wookey napsal(a):
> +++ Marek Vasut [2010-07-30 14:36 +0200]:
> > Dne P? 30. ?ervence 2010 14:36:09 Marek Vasut napsal(a):
> > > > > > support when the loon is used in this configuration. We have the
> > > > > > balloon_has() macro which is used for dealing with the different
> > > > > > builds of the board itself.
> > > > > > 
> > > > > > More obvious is using the CONFIG system to just enable this if
> > > > > > CUED_IO_BOARD is configured.
> > > > > 
> > > > > You can just disable the PCF driver if you want to save space. In
> > > > > case you won't have the CUED board connected, the driver will just
> > > > > fail to probe so it's ok I believe.
> > > > > 
> > > > > The macro could be extended, but do we want such a weird stuff in
> > > > > kernel?
> > > > 
> > > > That was how this list suggested we deal with the unprobe-able build
> > > > variation, I beleive.
> > > 
> > > We have loadable kernel modules ever since ... long time ago. Maybe
> > > that's more like the way to go.
> 
> But that makes it impossible to have a 'balloon' kernel that works on
> different hardware builds. So far we have been able to have one kernel
> build that works on all the different build variants by using that
> macro to avoid doing things which will hang forever when hardware is
> missing. Having to have a pile of different defconfigs for different
> build options helps no-one (and we don't like defconfigs anymore
> anyway :-)
> 
> More generic kernels is good, and ther has been much work on that sort
> of thing recently.

Maybe we should figure out a more generic way of doing this so it can be 
reusable on other machines too. Eric ? Maybe the Device tree would help here?
> 
> > > > > (especially if the driver can simply fail to probe).
> > > > 
> > > > No. Clearly anything that is probe-able should be probed. I thought
> > > > thtat this wouldn't be as I2C devices are pretty dumb, but I guess we
> > > > know its address here so we can try it and if no response then it's
> > > > not there.
> > > 
> > > Yes, this one is probeable.
> > 
> > Re-adding Eric as he was dropped. Cheers
> 
> OK. Just to be clear I am not objecting to any of the stuff Marek is
> posting here. It's great to have some of it upstream at last.
> 
> Once we have most of the basic support in we can consider the best
> ways of dealing with the odd configuration issues of balloon variants
> and add-on boards. (I'm not sure what you lot are going to think of
> our Samosa bus - hopefuly it's not too ugly these days as we did make
> it into a proper bus)
> 
> So, I am acking these patches.
> 
> Wookey