RE: Re: Isolation and time

["Dan Magenheimer" <dan.magenheimer@oracle.com>]

That's what I thought.  Sorry to belabour, but
that leads to one more question:

If one were to put an appropriately random CPU-only load
on every processor on domain0 (assuming domain0 runs
on all physical processors), then this would presumably
be sufficient to exercise a domain-under-test's time
synchronization, correct?

Thanks,
Dan

> -----Original Message-----
> From: xen-devel-bounces@lists.xensource.com
> [mailto:xen-devel-bounces@lists.xensource.com]On Behalf Of Keir Fraser
> Sent: Saturday, June 14, 2008 2:59 AM
> To: dan.magenheimer@oracle.com; Dave Winchell; Ben Guthro; xen-devel
> Subject: [Xen-devel] Re: Isolation and time
> 
> 
> On 14/6/08 03:20, "Dan Magenheimer" 
> <dan.magenheimer@oracle.com> wrote:
> 
> > But is there anything else?
> >
> > Suppose the credit scheduler were modified to
> > optionally schedule random "spurts" when the
> > sum of caps was less than the total available
> > CPU.  Would you then expect the results to be
> > essentially the same?
> 
> I wouldn't expect another domain's workload to affect the 
> test domain's time
> synchronisation except so far as the workload affects the domain's CPU
> demand over time. I imagine you could therefore simulate that 
> CPU demand
> process inside the scheduler. How hard that is presumably depends how
> accurate you want the simulation to be.
> 
>  -- Keir
> 
> 
> 
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xensource.com
> http://lists.xensource.com/xen-devel
>