[libvirt-users] dramatic performance slowdown due to THP allocation failure with full pagecache

[Daniel P. Berrange]

On Tue, Nov 14, 2017 at 11:03:30AM -0700, Blair Bethwaite wrote:
> On 14 November 2017 at 10:56, Daniel P. Berrange <berrange at redhat.com> wrote:
> > Oh well THP usage inside the guest is then not really anything todo with
> > virt, just a regular Linux questions, so not sure libvirt is the best
> > place to ask.
> 
> True, I just hoped you or one of the other devs might have some
> insight on reclaim behaviour that would provide a clue. I guess I'll
> try a linux list.
> 
> Any comment on this (which is virt related):
> (1) a related and possibly dumb question: in the case of a
> high-performance KVM where the guest is hugepage backed and pinned in
> host memory anyway, why do we still have a table based resolution for
> guest physical to host virtual address translation - couldn't this
> just be done by offset?

I don't know enough about KVM memory management to answer that. You probably
need to ask on the KVM mailing list to get the expert answer.


Regards,
Daniel
-- 
|: https://berrange.com      -o-    https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org         -o-            https://fstop138.berrange.com :|
|: https://entangle-photo.org    -o-    https://www.instagram.com/dberrange :|