Re: [PATCH] Add hypercall to mark superpages to improve performance

[Keir Fraser <keir.fraser@eu.citrix.com>]

On 28/04/2010 15:33, "Dave McCracken" <dcm@mccr.org> wrote:

> The current method of mapping hugepages/superpages in the hypervisor involves
> updating the reference counts of every page in the superpage.  This has proved
> to be a significant performance bottleneck.
> 
> This patch adds a pair of MMUEXT hypercalls to mark and unmark a superpage.
> Once the superpage is marked, the type is locked to writable page until a
> companion unmark is done.  When that superpage is subsequently mapped, only
> the first page needs to be reference counted.
> 
> There are checks when the superpage is marked and unmarked to make sure no
> individual page mappings have skewed the reference counts.

First of all, that changes the semantics of hugepages, since they can
subsequently *only* be mapped as superpages. I'm not sure that's a
restriction we want. Secondly, I don't really believe that the mark/unmark
hypercalls are race-free -- bearing in mind, that other mappings (superpage
or otherwise) can be constructed or deleted in parallel on other cpus.
Finally, does this really require new hypercalls? Could there not instead be
an always-enabled robust method for Xen to do superpage tracking?

 -- Keir