[PATCH 1/7] powerpc/mm: 64-bit 4k: use page-sized PMDs

[Benjamin Herrenschmidt]

> > Why do you want to create a virtual page table at the PMD level ? Also,
> > you are changing the geometry of the page tables which I think we don't
> > want. We chose that geometry so that the levels match the segment sizes
> > on server, I think it may have an impact with the hugetlbfs code (check
> > with David), it also was meant as a way to implement shared page tables
> > on hash64 tho we never published that.
> 
> The number of virtual page table misses were very high on certain loads.
> Cutting back to a virtual PMD eliminates most of that for the benchmark I
> tested, though it could still be painful for access patterns that are
> extremely spread out through the 64-bit address space.  I'll try a full
> 4-level walk and see what the performance is like; I was aiming for a
> compromise between random access and linear/localized access.

Let's get more numbers first then :-)

> Why does it need to match segment sizes on server?

I'm not sure whether we have a dependency with hugetlbfs there, I need
to check (remember we have one page size per segment there). For sharing
page tables that came from us using the PMD pointer as a base to
calculate the VSIDs. But I don't think we have plans to revive those
patches in the immediate future.

Cheers,
Ben.

> As for hugetlbfs, it merged easily enough with Becky's patches (you'll have
> to ask her when they'll be published).
> 
> -Scott