On Tue, 23 May 2017 11:48:54 -0300 Eduardo Habkost <ehabk...@redhat.com> wrote:
> On Tue, May 23, 2017 at 04:38:49PM +0200, Igor Mammedov wrote: > > Do the same as we did in commit > > (57924bcd8 numa: introduce machine callback for VCPU to node mapping) > > but only for incomplete mapping usecase, falling back to board > > provided default cpu to node mapping if user hasn't provided > > mapping for CPU explicitly. > > > > Signed-off-by: Igor Mammedov <imamm...@redhat.com> > > This breaks migration compatibility, doesn't it? I understand > this use case is not supported, but if are going to break stuff > for people using incomplete mappings (by rejecting the > configuration on a future release), I would prefer to break it > only once. it's firmware change and we generally don't care nor maintain firmware compat stuff the same as with above mentioned (57924bcd8 numa: introduce machine callback for VCPU to node mapping) > > > --- > > hw/core/machine.c | 3 --- > > 1 file changed, 3 deletions(-) > > > > diff --git a/hw/core/machine.c b/hw/core/machine.c > > index 964b75d..b8df15f 100644 > > --- a/hw/core/machine.c > > +++ b/hw/core/machine.c > > @@ -730,9 +730,6 @@ static void machine_numa_finish_init(MachineState > > *machine) > > g_string_append_printf(s, "%sCPU %d [%s]", > > s->len ? ", " : "", i, cpu_str); > > g_free(cpu_str); > > - > > - /* non mapped cpus used to fallback to node 0 */ > > - props.node_id = 0; > > } > > > > props.has_node_id = true; > > -- > > 2.7.4 > > >
