On Wed, 3 May 2017 11:59:35 -0300
Eduardo Habkost <ehabk...@redhat.com> wrote:
> On Wed, May 03, 2017 at 02:56:59PM +0200, Igor Mammedov wrote:
> [...]
> > @@ -378,14 +379,16 @@ void parse_numa_opts(MachineClass *mc)
> > * rule grouping VCPUs by socket so that VCPUs from the same socket
> > * would be on the same node.
> > */
> > + if (!mc->cpu_index_to_instance_props) {
> > + error_report("default CPUs to NUMA node mapping isn't
> > supported");
> > + exit(1);
> > + }
>
> This will make people trying to use -numa on unsupported machines
> see a misleading error message: instead of telling them that the
> machine doesn't support NUMA at all, the message seems to imply
> that NUMA may be supported and we just don't have default NUMA node
> mapping support.
>
> Probably a more generic "NUMA is not supported by this
> machine-type" message before even trying to parse -numa would be
> clearer. (I don't know if another patch in this series already
> does that.)
no, other places should error out with other error messages if it's
not supported.
