[OLPC-devel] Device tree

[Jim Gettys]

On Mon, 2006-08-14 at 07:40 -1000, Mitch Bradley wrote:
> [Changing name of thread...]
> 
> a) I think we need our own sub-architecture, sooner rather than later.  
> I could give a bunch of reasons, but I suspect that they are obvious to 
> everyone here, considering how little our hardware resembles a legacy PC.

True.  Marcelo, Dave, do you think this is a good route to go?  Will we
get any push back do you think from LKML or People That Matter (TM).
Certainly on ARM it is routine to get a sub-architecture; I don't know
how this is seen in the x86 part of the community.

> 
> b) Do we really need PCI autoconfiguration at this point?  The wiki 
> hardware spec doesn't mention the mini-PCI connector.  Can we just 
> pretend that it doesn't exist?  

Not clear the mini-pci will exist in production.  If it does, it will be
as pads only (no connector).

> That would remove a lot of variables 
> from the startup and suspend/resume sequence.  Removing unnecessary 
> variables is a Very Good Thing.

Our big set of variables is all the stuff/buses that can be on USB.  

> 
> c) I will go out on a limb and say that I could add rudimentary OF 
> device tree support to either or both of LinuxBIOS and a custom kernel 
> in a very short time.  Especially if the actual configuration is mostly 
> known.  No methods, just properties.  You may well ask: Why bother with 
> a device tree if the configuration is known?  The answer: Because when 
> we do the 2nd generation hardware and kernel, we will wish that we had.
> 

True.  I watched the iPAQ folks *try* to do this, though, and then not
follow through (so there wasn't sufficient information later).  If we're
going to do this, we'd better do it out of the box.
                                         - Jim

-- 
Jim Gettys
One Laptop Per Child