Re: [OpenBIOS] [PATCH 0/4] Changes for MorphOS

[BALATON Zoltan]

> I missed the reason you wanted the name changed; I commented at the time
> that for IEEE 1275, the name property at the root node is supposed to be
> the platform name. Why do you want it called "device-tree", which would
> certainly be incorrect?

To match Apple's implementation which is expected my at least one client. 
Sould I enclose it in CONFIG_PPC or some other ifdefs? Would that make it
better?

Regards,
BALATON Zoltan

[PATCH 0/4] Changes for MorphOS

[Tarl Neustaedter]

On 2014-Mar-8, 06:28 , BALATON Zoltan wrote:
> > I missed the reason you wanted the name changed; I commented at the time
> > that for IEEE 1275, the name property at the root node is supposed to be
> > the platform name. Why do you want it called "device-tree", which would
> > certainly be incorrect?
>
> To match Apple's implementation which is expected my at least one 
> client. Sould I enclose it in CONFIG_PPC or some other ifdefs? Would 
> that make it
> better?

Wow. Apple creates a name property for the root containing 
"device-tree"? That's bizarre.

In which case, assuming you're trying to emulate Apple, you'll have to 
live with it. I will note that Apple, IBM, Sun and Firmworks diverged on 
their implementations of IEEE 1275, to the degree that some card vendors 
had to have different FCodes for Sun systems vs Apple systems.

The name property in the root (like all other name properties) isn't 
supposed to be meaningful to code, just useful for printing. Anything 
where code is supposed to match drivers to nodes should use the 
"compatible" property. But if the client is written to use the name 
property, we have to live with it and supply what the client needs.

I don't know if you'd want to include that change under CONFIG_PPC or 
perhaps under some flag for Apple emulation - IBM's PPC systems may not 
follow Apple's lead on that.

[PATCH 0/4] Changes for MorphOS

[Mark Cave-Ayland]

On 08/03/14 16:28, Tarl Neustaedter wrote:
> Wow. Apple creates a name property for the root containing
> "device-tree"? That's bizarre.
>
> In which case, assuming you're trying to emulate Apple, you'll have to
> live with it. I will note that Apple, IBM, Sun and Firmworks diverged on
> their implementations of IEEE 1275, to the degree that some card vendors
> had to have different FCodes for Sun systems vs Apple systems.
>
> The name property in the root (like all other name properties) isn't
> supposed to be meaningful to code, just useful for printing. Anything
> where code is supposed to match drivers to nodes should use the
> "compatible" property. But if the client is written to use the name
> property, we have to live with it and supply what the client needs.
>
> I don't know if you'd want to include that change under CONFIG_PPC or
> perhaps under some flag for Apple emulation - IBM's PPC systems may not
> follow Apple's lead on that.

Yeah; I suspect this is going to have to be one behaviour that we are 
going to have to end up emulating for compatibility purposes.

I think the best solution is going to be to move the default device name 
from forth/device/tree.fs into the individual per-architecture 
arch/*/tree.fs files and then customise the PPC one as appropriate.

ATB,

Mark.