Re: [Qemu-devel] [RFC PATCH v2 9/9] spapr: CPU hot unplug support

[Bharata B Rao <bharata@linux.vnet.ibm.com>]

On Wed, Mar 16, 2016 at 04:27:04PM +1100, David Gibson wrote:
> On Fri, Mar 11, 2016 at 10:24:38AM +0530, Bharata B Rao wrote:
> > Remove the CPU core device by removing the underlying CPU thread devices.
> > Hot removal of CPU for sPAPR guests is achieved by sending the hot unplug
> > notification to the guest. Release the vCPU object after CPU hot unplug so
> > that vCPU fd can be parked and reused.
> > 
> > Signed-off-by: Bharata B Rao <bharata@linux.vnet.ibm.com>
> > ---
> >  hw/ppc/spapr.c                  | 21 ++++++++++
> >  hw/ppc/spapr_cpu_core.c         | 86 +++++++++++++++++++++++++++++++++++++++++
> >  include/hw/ppc/spapr.h          |  1 +
> >  include/hw/ppc/spapr_cpu_core.h | 12 ++++++
> >  4 files changed, 120 insertions(+)
> > 
> > diff --git a/hw/ppc/spapr.c b/hw/ppc/spapr.c
> > index 822c87d..b1e9ba2 100644
> > --- a/hw/ppc/spapr.c
> > +++ b/hw/ppc/spapr.c
> > @@ -2345,7 +2345,12 @@ static void spapr_machine_device_plug(HotplugHandler *hotplug_dev,
> >  
> >          spapr_memory_plug(hotplug_dev, dev, node, errp);
> >      } else if (object_dynamic_cast(OBJECT(dev), TYPE_SPAPR_CPU_CORE)) {
> > +        /*
> > +         * TODO: Move this check to pre_plug handler at which point
> > +         * spapr_core_release() won't be necessary.
> > +         */
> >          if (!smc->dr_cpu_enabled && dev->hotplugged) {
> > +            spapr_core_release(dev);
> >              error_setg(errp, "CPU hotplug not supported for this machine");
> >              return;
> >          }
> 
> This hunk doesn't look like its related to unplug.  Did it belong in
> another patch?

Yes, it actually belongs to hot-plug but the whole infrastructure to release
the core and associated threads get introduced in this patch, hence put
this hunk here.

However, if pre_plug is the way to go forward, we woudn't need this
altogether as roll-back is much easier from there than from here.

Regards,
Bharata.