Re: [PATCH] fix/improve generic page table walker

[Martin Schwidefsky <schwidefsky@de.ibm.com>]

On Mon, 16 Mar 2009 13:36:54 +0100
Nick Piggin <npiggin@suse.de> wrote:

> > With the page table folding "3 levels removed from the bottom" doesn't
> > tell me much since there is no real representation in hardware AND in
> > memory for the missing page table levels. So the only valid meaning of
> > a pgd_t is that you have to use pud_offset, pmd_offset and pte_offset
> > to get to a pte. If I do the page table folding at runtime or at
> > compile time is a minor detail.  
> 
> I don't know if it would be helpful to you, but I solve a similar
> kind of problem in the lockless radix tree by encoding node height
> in the node itself. Maybe you could use some bits in the page table
> pointers or even in the struct pages for this.

That is what I already do: there are two bits in the region and segment
table entries that tell me at what level I am (well actually it is the
hardware definition that requires me to do that and I just make use of
it). The page table primitives (pxd_present, pxd_offset, etc) look at
these bits and then do the right thing.
What is killing me is the pgd++/pud++ operation. If there is only a 2
or 3 level page table the pointer increase may not happen. This is done
by a correct end address for the walk. 

-- 
blue skies,
   Martin.

"Reality continues to ruin my life." - Calvin.