Re: [PATCH v12 1/7] s390x/cpu topology: Creating CPU topology device

[Pierre Morel]

On 12/1/22 10:08, Thomas Huth wrote:
> On 29/11/2022 18.42, Pierre Morel wrote:
> > We will need a Topology device to transfer the topology
> > during migration and to implement machine reset.
> >
> > The device creation is fenced by s390_has_topology().
> >
> > Signed-off-by: Pierre Morel <pmorel@linux.ibm.com>
> > ---
> ...
> > diff --git a/hw/s390x/s390-virtio-ccw.c b/hw/s390x/s390-virtio-ccw.c
> > index 2e64ffab45..973bbdd36e 100644
> > --- a/hw/s390x/s390-virtio-ccw.c
> > +++ b/hw/s390x/s390-virtio-ccw.c
> > @@ -44,6 +44,7 @@
> >   #include "hw/s390x/pv.h"
> >   #include "migration/blocker.h"
> >   #include "qapi/visitor.h"
> > +#include "hw/s390x/cpu-topology.h"
> >   static Error *pv_mig_blocker;
> > @@ -102,6 +103,24 @@ static void s390_init_cpus(MachineState *machine)
> >       }
> >   }
> > +static DeviceState *s390_init_topology(MachineState *machine, Error 
> > **errp)
> > +{
> > +    DeviceState *dev;
> > +
> > +    dev = qdev_new(TYPE_S390_CPU_TOPOLOGY);
> > +
> > +    object_property_add_child(&machine->parent_obj,
> > +                              TYPE_S390_CPU_TOPOLOGY, OBJECT(dev));
> > +    object_property_set_int(OBJECT(dev), "num-cores",
> > +                            machine->smp.cores * 
> > machine->smp.threads, errp);
>
> I wonder what will happen if we ever support multithreading on s390x 
> later? ... won't this cause some oddities when migrating older machines 
> types with smp.threads > 1 later? Maybe we should prohibit to enable the 
> CPU topology instead if a user tried to use threads > 1 with an older 
> machine type?

Yes, right, I forgot to change this back.
Anyway it has no sens for new machine which prohibit smp.threads > 1 
neither.

I change this by returning an error in case we have smp.threads > 1

Thanks.

Pierre

>   Thomas
>
>
> > +    object_property_set_int(OBJECT(dev), "num-sockets",
> > +                            machine->smp.sockets, errp);
> > +
> > +    sysbus_realize_and_unref(SYS_BUS_DEVICE(dev), errp);
> > +
> > +    return dev;
> > +}

--
Pierre Morel
IBM Lab Boeblingen