On 11/07/2012 08:22 AM, Peter Krempa wrote:
> Lately there were a few reports of the output of the virsh nodeinfo
> command being inaccurate. This patch ties to avoid that by checking if
s/ties/tries/
> the topology actualy makes sense. If it doesn't we then report a
s/actualy/actually/
> synthetic topology that indicates to the user that the host capabilities
> should be checked for the actual topology.
> ---
> src/nodeinfo.c | 37 +++++++++++++++++++++++++++++++------
> 1 file changed, 31 insertions(+), 6 deletions(-)
You've convinced me that we need this; and we have already documented
that nodes=1 is a hint that the user must verify with the capability XML
for accurate results anyway. ACK if you can answer one question:
> @@ -531,6 +539,23 @@ done:
> goto cleanup;
> }
>
> + /* Now check if the topology makes sense. There are machines that don't
> + * expose their real number of nodes or for example the AMD Bulldozer
> + * architecture that exposes their Clustered integer core modules as both
> + * threads and cores. This approach throws off our detection. Unfortunately
> + * the nodeinfo structure isn't designed to carry the full topology so
> + * we're going to lie about the detected topology to notify the user
> + * to check the host capabilities for the actual topology. */
> + if ((nodeinfo->nodes *
> + nodeinfo->sockets *
> + nodeinfo->cores *
> + nodeinfo->threads) != (nodeinfo->cpus + offline)) {
> + nodeinfo->nodes = 1;
> + nodeinfo->sockets = nodeinfo->cpus;
> + nodeinfo->cores = 1;
Would it be any better to swap these and expose sockets == 1 and cores
== cpus when we fake things?
--
Eric Blake eblake@xxxxxxxxxx +1-919-301-3266
Libvirt virtualization library http://libvirt.org
Attachment:
signature.asc
Description: OpenPGP digital signature
-- libvir-list mailing list libvir-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/libvir-list