[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-devel] [OSSTEST PATCH 1/2] TestSupport.pm: allow creating vNUMA enabled HVM guest configs



On Fri, Oct 02, 2015 at 02:02:15PM +0200, Dario Faggioli wrote:
> On Fri, 2015-10-02 at 12:32 +0100, Wei Liu wrote:
> > On Fri, Oct 02, 2015 at 01:17:31AM +0200, Dario Faggioli wrote:
> 
> > > diff --git a/Osstest/TestSupport.pm b/Osstest/TestSupport.pm
> > > index 2b67e32..bf8bd56 100644
> > > --- a/Osstest/TestSupport.pm
> > > +++ b/Osstest/TestSupport.pm
> > > @@ -1709,6 +1709,34 @@ sub prepareguest_part_xencfg ($$$$$) {
> 
> > > +        # Let's put all vnodes on pnode 0, as we don't even know
> > > (for now)
> > > +        # whether or not the host has more NUMA nodes than that!
> > > +        @vnuma_pnode = (0) x $vnuma_nodes;
> > > +
> > 
> > The thing I worry about putting everything in node 0 is that if we
> > somehow has hardware that has several nodes but each node has
> > relative
> > small amount of ram it would cause this test case to fail.
> > 
> ISWYM.
> 
> > In the current arrangement, HVM guest gets 5G of ram if OSSTest
> > determines the host has enough (2x5G + slack) ram for it.  So vnuma
> > test
> > case will fail on a box with 4 nodes but has 16G of ram. Is such
> > hardware configuration not very likely to exist?
> > 
> Well, 16GB is a rather small amount of RAM for a box with 4 nodes, IMO.
> 
> However, I think your point is valid... I think I can, in case
> vnuma_nodes is provided, overrun $ram_mb with something more likely to
> fit on one physical node?
> 

Missed this one, sorry.

I'm not sure what exact thing you proposed. But I think it would be
valuable to have the guest ram larger than 4GB. That would certainly
catch many bugs that involves memory layout changes, especially for
feature like vnuma.

> Unfortunately, it's not something we can check precisely, because, in
> Xen, there is (yet) no mechanism to know, in case we have XX free RAM
> (apart from parsing the console output after issuing `xl debug-key u',
> which is really really unpractical!), how many of it actually comes
> from each NUMA ndoe... but I can apply a bit of guessing.
> 
> What do you think?
> 
> Regards,
> Dario
> -- 
> <<This happens because I choose it to happen!>> (Raistlin Majere)
> -----------------------------------------------------------------
> Dario Faggioli, Ph.D, http://about.me/dario.faggioli
> Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK)
> 



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


 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.