[PATCH 6/10] ppc64: Create a fake flat device tree on iSeries

[Milton Miller]

On Aug 10, 2005, at 3:32 AM, David Gibson wrote:

> On Wed, Aug 10, 2005 at 02:16:15AM -0500, Milton Miller wrote:
>>
>> 5) If you are going to statically allocate this for iSeries (as
>> opossed to say a klimit bump), then the only thing in dynamic is
>> setting the boot cpuid.  May as well use a static initializer.
>>
>> Today this would add over 2 pages of bss for a "combined" kernel.
>
> I think you'll find there are patches coming making more stuff
> dynamic.  Virtual devices, in particular.

Sorry, wasn't trying to imply making the whole tree static, just
the header.   But I would rather see it allocated at klimit ...


>
>> Putting the
>> device data at klimit would allow the checks for overflow to be
>> deferred until after its created (look for 2 nuls in string table
>> -- the size is propnum(""), and placing data at after the reserve
>> map and string table and either compare to total size or set
>> totalsize according to what is used).
>
> Hrm.. actually.. this may be too irritatingly inflexible, but since
> the iseries device tree is pretty limited in what it contains, it
> would potentially be possible to statically initialize a string table
> containing every property name iSeries might ever use.  In fact, doing
> it that way you could generate the skeleton of the device tree with
> dtc (asm output mode), leaving some space in the structure section,
> then add in the dynamic structure at runtime.  (I'd need to add
> something to dtc to allow you to force a string into the string
> section even if it's not used in the tree, of course, but that should
> be easy)

My other generator does this very thing.  The strings are allocated
statically.

While discussing embedded with panto on #mklinux, another approach
is to say "here's the max", put the strings at the end, and allocate
new strings down moving the start as we add strings.   Makes sense for
fw where there is a fixed allocation.  End of blob processing can
then memmove the string table and cut total size down.

Even without the knowing the max size realloc could move strings down
if a space check is added before extending the dt blob.

>
> -- 
> David Gibson			| I'll have my music baroque, and my code
> david AT gibson.dropbear.id.au	| minimalist, thank you.  NOT _the_ 
> _other_
> 				| _way_ _around_!
> http://www.ozlabs.org/people/dgibson
>