[PATCH 02/11] arm: use generic infrastructure for early params

[Rob Landley]

On Wednesday 06 January 2010 16:31:00 Jeremy Kerr wrote:
> Russell,
>
> > This is something which should be separate from the rest of the DT
> > patch - could we have this as a patch which can be applied to the
> > current kernel tree please?
>
> Sure, patch coming.
>
> The first three patches of this series (prefixed with arm:, rather than
> arm- dt:) are all generic arm patches that don't specifically add device
> tree support, would you like the other two as well?
>
> I'm in the process of rebasing my work onto a merge or your and Grant
> Likely's test-devicetree branch, so future patches should be easier to work
> with.
>
> Cheers,

What's the current status on the device tree for arm stuff?

I'm going through some of my todo bookmarks and stumbled across 
http://lwn.net/Articles/367752/ and I'd like to poke at it under qemu.

I've been patching the kernel kconfig stuff to let me plug various different arm 
processors into a versatile board (because it's convinced that you can't 
possibly have an armv6l versatilepb and qemu's -cpu option makes that easy).  
Feeding in a device tree (or better yet teaching qemu to generate one) sounds 
like a better long-term approach...

Rob
-- 
Latency is more important than throughput. It's that simple. - Linus Torvalds