This is an archived copy of the Xen.org mailing list, which we have preserved to ensure that existing links to archives are not broken. The live archive, which contains the latest emails, can be found at http://lists.xen.org/
Home Products Support Community News


Re: [Xen-devel] Re: [Xen-staging] [xen-unstable] Added some more fields

To: Keir Fraser <Keir.Fraser@xxxxxxxxxxxx>
Subject: Re: [Xen-devel] Re: [Xen-staging] [xen-unstable] Added some more fields to host_cpu.
From: Alex Williamson <alex.williamson@xxxxxx>
Date: Fri, 02 Mar 2007 13:55:07 -0700
Cc: Jimi Xenidis <jimix@xxxxxxxxxxxxxx>, xen-devel <xen-devel@xxxxxxxxxxxxxxxxxxx>, ewan@xxxxxxxxxxxxx, Hollis Blanchard <hollisb@xxxxxxxxxx>
Delivery-date: Fri, 02 Mar 2007 12:54:22 -0800
Envelope-to: www-data@xxxxxxxxxxxxxxxxxx
In-reply-to: <C20E3D8D.32B5%Keir.Fraser@xxxxxxxxxxxx>
List-help: <mailto:xen-devel-request@lists.xensource.com?subject=help>
List-id: Xen developer discussion <xen-devel.lists.xensource.com>
List-post: <mailto:xen-devel@lists.xensource.com>
List-subscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=unsubscribe>
Organization: HP OSLO R&D
References: <C20E3D8D.32B5%Keir.Fraser@xxxxxxxxxxxx>
Sender: xen-devel-bounces@xxxxxxxxxxxxxxxxxxx
On Fri, 2007-03-02 at 20:46 +0000, Keir Fraser wrote:
> On 2/3/07 20:39, "Alex Williamson" <alex.williamson@xxxxxx> wrote:
> >   On ia64, dom0 doesn't automatically get vcpus for each physical cpu,
> > so the first problem is that we're not going to have a /proc/cpuinfo
> > entry for every cpu in self.cpus.keys.  I think it's likely x86 could
> > run into this problem too if a cpu was hotplugged or booted with the
> > dom0_max_vcpus options.
> We have indeed hit this problem and I put a patch in this afternoon to
> duplicate cpu0's info for any non-existent cpu. Given that the dom0 cpus
> could be migrating around on arbitrary physical cpus (even across the
> multiple CPUID invocations that the kernel will have made to build the
> information for a single 'cpu' in /proc!) this is fine -- x86 multiprocessor
> systems are supposed to be symmetric (homogeneous CPUs down to the same
> stepping in some cases) anyway.

   Cool, I see it in staging.  This is the approach I used to work
around the problem temporarily, but what happens if cpu0 is hot
un-plugged?  ISTR x86 Linux doesn't support cpu0 hotplug, but on ia64 we
can hotplug cpu0.  I'd guess powerpc could too.  Thanks,


Alex Williamson                             HP Open Source & Linux Org.

Xen-devel mailing list