Re: [Xen-devel] [PATCH 04/12] arm: parse modules from DT during early boot.

[Stefano Stabellini <stefano.stabellini@xxxxxxxxxxxxx>]

On Thu, 29 Nov 2012, Ian Campbell wrote:
> On Thu, 2012-11-29 at 17:05 +0000, Tim Deegan wrote:
> > > +            cell = (const u32 *)prop->data;
> > > +            device_tree_get_reg(&cell, address_cells, size_cells,
> > > +                                &mod->start, &size);
> > 
> > This get_reg returns a start + size -- can/should we encode the module
> > as one of these rather than encdong start + end separately and
> > discarding the 'size' fields?
> 
> Interesting thought, I'm not enough of a DTB guru to know what the right
> way to express this is (CCing Stefano :-))
> 
> This is trying to parse
> / {
>       chosen { 
>               module@1 {
>                       start = 0x80000000;
>                       end = 0x2000;
>               }
>       }
> which is roughtly equivalent to how Linux bootloaders pass in initrds
> (although the name etc differ)
> 
> I suspect using device_tree_get_reg as things stands is just plain
> wrong, since the above things are not actually regs.
> 
> However you might be right that this should be expressed as
> 
> / {
>       chosen { 
>               module@1 {
>                       address = <0x80000000 0x2000>;
>               }
>       }
> 
> and then I think using device_tree_get_reg would be correct.
> 
> Stefano -- does that make sense? is "address = < ... >" allowed or does
> the thing have to be called reg?

I don't think we should use device_tree_get_reg to parse something that
is not a reg. If we want a reg then we should just call the property
"reg" (I am in favor of that).

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel