bhyve: Make most I/O port handling specific to amd64

[markj]

Summary

• The qemu_fwcfg interface, as implemented, is I/O port-based, but QEMU implements an MMIO interface that we'll eventually want to port for arm64.
• Retain support for I/O space PCI BARs, simply treat them like MMIO BARs for most purposes, similar to what the arm64 kernel does. Such BARs are created by virtio devices.

[jhb]

usr.sbin/bhyve/pci_emul.c

539

I would maybe structure this as:

#ifdef __amd64__
static int
pci_emul_io_handler(... inout args)
{
   inout version
}
#else
static int
pci_emul_io_handler(... mem args)
{
  mr version
}
#endif

640

I think I'd almost rather do a bit more duplication here instead as I would find it more readable I think:

case PCIBAR_IO:
#ifdef __amd64__
     bzero(&iop, sizeof(struct inout_port));
     ...
#else
     bzero(&mr, sizeof(struct mem_range));
#endif
     break;

One thing that would help perhaps to reduce duplication would be to first move the pe_baraddr calls out of the case statement entirely as they are duplicated for all cases.

[markj]

Apply jhb's suggestions.