Patches added to powerpc.git master and powerpc-next branches

[Kumar Gala]

On Apr 19, 2008, at 7:18 AM, Paul Mackerras wrote:
> Kumar Gala writes:
>
>> [POWERPC] 85xx: Add support for relocatble kernel (and booting at  
>> non-
>> zero)
>
> Should be OK to go though probably not in the first batch.  I want to
> look through it carefully again since it's touching code that is
> common to a lot of platforms.  The patch description looks reasonable
> as to what it explains but the english could use a touch-up here or
> there (e.g. "a physical address that is module 256M" doesn't actually
> mean anything; "a physical address that is 0 modulo 256M" would be
> correct).

Ok.  I'll have my grammar police review it and repost.

>> [POWERPC] Port fixmap from x86 and use for kmap_atomic
>
> Once again I want to go through it carefully and understand how it all
> works, and in particular understand things like the way it ensures
> that the base address for the kmap region is aligned to a 4M boundary
> so all the kmap ptes are in a single page (assuming it does :).

I'm not clear on why this is needed.  I can see value in having the  
PGD entry in place but the pte PAs would be changing all the time so I  
don't see what benefit there is in keeping them in one page.

For normal kmap ptes we cover things in one PGD via:

#define PKMAP_BASE      ((FIXADDR_BOOT_START - PAGE_SIZE*(LAST_PKMAP +  
1)) & PMD_MASK)

The kmap_atomic ptes do NOT have a similar guarantee as they are  
controlled via fixmap.  However we do "pre-allocate" the PGD via  
calling map_page() on all fixmap addresses.

- k