Re: [Xen-devel] [PATCH 03 of 10 v3] xl: add more NUMA information to `xl info -n'

[Ian Campbell <Ian.Campbell@xxxxxxxxxx>]

On Fri, 2012-07-06 at 13:00 +0100, Dario Faggioli wrote:
> On Fri, 2012-07-06 at 12:37 +0100, Ian Campbell wrote: 
> > > xl: add more NUMA information to `xl info -n'
> > > 
> > > So that the user knows how much memory there is on each node and
> > > how far they are from each others.
> > > 
> > > Signed-off-by: Dario Faggioli <dario.faggioli@xxxxxxxxxx>
> > > Acked-by: Ian Campbell <ian.campbell@xxxxxxxxxx>
> > 
> > On my single node system this produces:
> >         cpu_topology           :
> >         cpu:    core    socket     node
> >           0:       0        0        0
> >           1:       1        0        0
> >           2:       2        0        0
> >           3:       3        0        0
> >         numa_info              :
> >         node:    memsize    memfree    distances
> >            0:      4608       3083      10
> > 
> > Is "distances" here right? 
> >
> It looks the same here. While with two nodes it gives something like the
> below:
> 
>      0:  1:
>   0: 10, 20
>   1: 20, 10
> 
> More important, forgetting about the '10' do you want me to suppress
> that part of the output if there's only one node? I did not think at it
> before, but it might make sense. Of course, I can also do this with a
> follow-up patch, just let me know.

Oh, it's a decimal number, for some reason I was reading it in a bitmap
or a "distance per column" way (i.e. as dist[0]=0 and dist[1]=1), which
is non-sensical now I think about it

> 
> > I'd have expected only a single 0 (distance
> > from self)?
> > 
> Honestly, I've not looked at where it comes from... I think I've seen
> something similar in Linux as well (although I might be wrong).
> 
> Do you want me to investigate?

It's ok, was just me being dumb ;-)

I'll commit this shortly, along with the patches up until (and including
#7)

Ian


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