[PATCH v4 2/2] bhyve: support 'isa' controller for LPC

[Roman Bogorodskiy]

  Daniel P. Berrangé wrote:

> On Wed, Jul 15, 2020 at 07:25:45PM +0400, Roman Bogorodskiy wrote:
> > Support modeling of the 'isa' controller for bhyve. User can manually
> > define any PCI slot for the 'isa' controller, including PCI slot 1,
> > but other devices are not allowed to use this address.
> > 
> > When domain configuration requires the 'isa' controller to be present,
> > automatically add it on domain post-parse stage.
> > 
> > Now, as this controller is always available when needed, it's not
> > necessary to implicitly add it to the bhyve command line, so remove
> > bhyveBuildLPCArgStr().
> > 
> > Also, make bhyveDomainDefNeedsISAController() static as it's no longer
> > used outside of bhyve_domain.c.
> > 
> > As more than one ISA controller is not supported by bhyve,
> > and multiple controllers with the same index are forbidden,
> > so forbid ISA controllers with non-zero index for bhyve.
> > 
> > Signed-off-by: Roman Bogorodskiy <bogorodskiy at gmail.com>
> 
> > diff --git a/src/bhyve/bhyve_device.c b/src/bhyve/bhyve_device.c
> > index fc52280361..52a055f205 100644
> > --- a/src/bhyve/bhyve_device.c
> > +++ b/src/bhyve/bhyve_device.c
> > @@ -46,10 +46,16 @@ bhyveCollectPCIAddress(virDomainDefPtr def G_GNUC_UNUSED,
> >          if (addr->slot == 0) {
> >              return 0;
> >          } else if (addr->slot == 1) {
> > -            virReportError(VIR_ERR_INTERNAL_ERROR, "%s",
> > -                           _("PCI bus 0 slot 1 is reserved for the implicit "
> > -                             "LPC PCI-ISA bridge"));
> > -            return -1;
> > +            if (!(device->type == VIR_DOMAIN_DEVICE_CONTROLLER &&
> > +                  device->data.controller->type == VIR_DOMAIN_CONTROLLER_TYPE_ISA)) {
> > +                 virReportError(VIR_ERR_INTERNAL_ERROR, "%s",
> > +                                _("PCI bus 0 slot 1 is reserved for the implicit "
> > +                                  "LPC PCI-ISA bridge"));
> > +                 return -1;
> > +            } else {
> > +                /* We reserve slot 1 for LPC in bhyveAssignDevicePCISlots(), so exit early */
> > +                return 0;
> > +            }
> 
> IIUC, this series makes it possible to put the TPC in a different
> slot, so does it still make sense to forbid use of slot 1 as a
> hardcoded rule ?

IIRC, the idea behind that is to give some time window for users to
allow moving guests from the new version to the old one. If we allow to
use slot 1, it won't be possible to move the guest to the old libvirt as
it will complain slot 1 should be used only for LPC.

> 
> Regards,
> Daniel
> -- 
> |: https://berrange.com      -o-    https://www.flickr.com/photos/dberrange :|
> |: https://libvirt.org         -o-            https://fstop138.berrange.com :|
> |: https://entangle-photo.org    -o-    https://www.instagram.com/dberrange :|
> 

Roman Bogorodskiy
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 488 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/libvir-list/attachments/20200727/66d6244b/attachment-0001.sig>