patch: allow dts include files to have nodes

[ron minnich]

per IRC discussions.

ron

[Myles Watson]

It didn't work for me.
> -----Original Message-----
> From: [e-mail] [mailto:[e-mail]]
> On Behalf Of ron minnich
> Sent: Thursday, November 13, 2008 10:51 AM
> To: Coreboot
> Subject: [coreboot] patch: allow dts include files to have nodes
>
> per IRC discussions.
>
> ron

It didn't work for me.  I put the pnp devices in the SIO dts, and it says it
can't read the input tree (unfortunately it gives a line number I didn't
change in the error.) It works fine on the old tree with no nodes.

Thanks,
Myles

[ron minnich]

On Thu, Nov 13, 2008 at 10:19 AM, Myles Watson  wrote:
> It didn't work for me.
> > -----Original Message-----
> > From: [e-mail] [mailto:[e-mail]]
> > On Behalf Of ron minnich
> > Sent: Thursday, November 13, 2008 10:51 AM
> > To: Coreboot
> > Subject: [coreboot] patch: allow dts include files to have nodes
> >
> > per IRC discussions.
> >
> > ron
>
> It didn't work for me.  I put the pnp devices in the SIO dts, and it says it
> can't read the input tree (unfortunately it gives a line number I didn't
> change in the error.) It works fine on the old tree with no nodes.

it's not going to do much yet. This is to allow subnodes to be in the
tree syntactically. There is zero processing occuring in the
flatten_tree at this point.

I just built a geode target with it. Are you building against your
modified tree?

This is a step-by-step. It should work on unmodified trees -- it did for me.

I would suggest that you start with a clean tree and let's go through
your ideas/changes one by one. I think you are in the 'diverged too
far' state.

ron

[Myles Watson]

On Thu, Nov 13, 2008 at 11:45 AM, ron minnich  wrote:
> On Thu, Nov 13, 2008 at 10:19 AM, Myles Watson  wrote:
> > It didn't work for me.
> > > -----Original Message-----
> > > From: [e-mail] [mailto:
>
> [e-mail]]
> > > On Behalf Of ron minnich
> > > Sent: Thursday, November 13, 2008 10:51 AM
> > > To: Coreboot
> > > Subject: [coreboot] patch: allow dts include files to have nodes
> > >
> > > per IRC discussions.
> > >
> > > ron
> >
> > It didn't work for me.  I put the pnp devices in the SIO dts, and it says
>
> it
> > can't read the input tree (unfortunately it gives a line number I didn't
> > change in the error.) It works fine on the old tree with no nodes.

Sorry,

I didn't see this response.

it's not going to do much yet. This is to allow subnodes to be in the
> tree syntactically. There is zero processing occuring in the
> flatten_tree at this point.

It doesn't parse my tree if I put in an extra node in a dts.
> I just built a geode target with it. Are you building against your
> modified tree?

Yes, but it works fine until I add a node.
> This is a step-by-step. It should work on unmodified trees -- it did for
> me.

It did for me too.
> I would suggest that you start with a clean tree and let's go through
> your ideas/changes one by one. I think you are in the 'diverged too
> far' state.

I don't think the PNP patch I sent is really that big, and it is separate
from everything else.  How would you suggest I break it up?

Thanks,
Myles

[Myles Watson]

On Thu, Nov 13, 2008 at 1:48 PM, Myles Watson  wrote:
> On Thu, Nov 13, 2008 at 11:45 AM, ron minnich  wrote:
> > On Thu, Nov 13, 2008 at 10:19 AM, Myles Watson  wrote:
> > > It didn't work for me.
> > > > -----Original Message-----
> > > > From: [e-mail] [mailto:
> >
> > [e-mail]]
> > > > On Behalf Of ron minnich
> > > > Sent: Thursday, November 13, 2008 10:51 AM
> > > > To: Coreboot
> > > > Subject: [coreboot] patch: allow dts include files to have nodes
> > > >
> > > > per IRC discussions.
> > > >
> > > > ron
> > >
> > > It didn't work for me.  I put the pnp devices in the SIO dts, and it
> >
> > says it
> > > can't read the input tree (unfortunately it gives a line number I didn't
> > > change in the error.) It works fine on the old tree with no nodes.

After lunch and a make clean, it now generates the expected statictree.c.
As you said, it doesn't do much, but it adds a dev_(null) in the device list
where the new node would be.

I don't see how it solves the basic problem of PNP device allocation and dts
parameter passing, but I see how it could make the top-level dts prettier.
Can we wait on beautification until we have it all the way working?

Thanks,
Myles

[Marc Jones]

----- Original Message ----
> From: ron minnich 
> To: Coreboot 
> Sent: Thursday, November 13, 2008 10:51:06 AM
> Subject: [coreboot] patch: allow dts include files to have nodes
>
> per IRC discussions.
>
> ron

I think that this is what we want to do. I would like to know what Myles thinks.

Acked-by: Marc Jones

Marc

[ron minnich]

On Thu, Nov 13, 2008 at 10:25 AM, Marc Jones  wrote:
> I think that this is what we want to do. I would like to know what Myles thinks.
>
> Acked-by: Marc Jones
>
> Marc

can somebody confirm a build/boot with this change?

thanks

ron

[Myles Watson]

On Thu, Nov 13, 2008 at 11:45 AM, ron minnich  wrote:
> On Thu, Nov 13, 2008 at 10:25 AM, Marc Jones  wrote:
> > I think that this is what we want to do. I would like to know what Myles
>
> thinks.
> > Acked-by: Marc Jones
> >
> > Marc
>
> can somebody confirm a build/boot with this change?

This is copied from my mail at 11:19.  Did it not make it through?

It didn't work for me.  I put the pnp devices in the SIO dts, and it says it
can't read the input tree (unfortunately it gives a line number I didn't
change in the error.) It works fine on the old tree with no nodes.

Thanks,
Myles

[ron minnich]

send my your sio dts

ron