Re: [PATCH 1 of 3 v5/leftover] libxl: enable automatic placement of guests on NUMA nodes [and 1 more messages]

[Dario Faggioli <raistlin@linux.it>]

[-- Attachment #1.1: Type: text/plain, Size: 1500 bytes --]

On Thu, 2012-07-19 at 20:37 +0200, Andre Przywara wrote:
> > But it's OK, we don't need to panic.  We just need a safety catch
> > which stops this algorithm running in situations where it won't work.
> > Following a discussion with Dario, AIUI he plans to implement a rule
> > that it will bail on systems with more than 8 NUMA nodes.
> >
> > That is releasable in 4.2 and ought to be satisfactory for you ?
> 
I did right that. Honestly, I computed the exact amount of steps the
algorithm needs in the worst case (i.e., searching all the combinations
of all the sizes) for 8 and 16 nodes. 8 nodes means 256 steps, 16 nodes
means 65536, so I think both could work, especially after the
optimization I put in the version I just released (it does not uses that
large array of combinations and does not need to always perform all the
steps any longer).

However, as agreed, I put an "if (nr_nodes <= 8)" there... If a 16 nodes
machine will ever appear, we'll decide what to do.

> Absolutely, that's fine. I just wanted to make sure that we don't drop 
> the patches at all because of the theoretical fear of a large NUMA box 
> lingering somewhere around the corner...
> 
:-)

Thanks and Regards,
Dario

-- 
<<This happens because I choose it to happen!>> (Raistlin Majere)
-----------------------------------------------------------------
Dario Faggioli, Ph.D, http://retis.sssup.it/people/faggioli
Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK)


[-- Attachment #1.2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

[-- Attachment #2: Type: text/plain, Size: 126 bytes --]

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel