Re: [PATCH] qemu_domain_address: fix CCW virtio-mem hotplug

On Tue, Mar 18, 2025 at 09:53:22AM +0100, Boris Fiuczynski wrote: > Since commit f23f8ff91a virtio-mem supports also CCW. When hotplugging a > virtio-mem device with a CCW address results in a PCI device getting > attached. The method qemuDomainAssignMemoryDeviceSlot is only > considering PCI as address type and overwriting the CCW address. Adding > support for address type CCW. > > Signed-off-by: Boris Fiuczynski <fiuczy(a)linux.ibm.com&gt; > --- > src/qemu/qemu_domain_address.c | 16 +++++++++++++++- > 1 file changed, 15 insertions(+), 1 deletion(-) > > diff --git a/src/qemu/qemu_domain_address.c b/src/qemu/ > qemu_domain_address.c > index d38983bf63..b0289c2337 100644 > --- a/src/qemu/qemu_domain_address.c > +++ b/src/qemu/qemu_domain_address.c > @@ -3067,6 +3067,7 @@ qemuDomainAssignMemoryDeviceSlot(virDomainObj *vm, >                                  virDomainMemoryDef *mem) > { >     g_autoptr(virBitmap) slotmap = NULL; > +    virDomainCCWAddressSet *ccwaddrs = NULL; >     virDomainDeviceDef dev = {.type = > VIR_DOMAIN_DEVICE_MEMORY, .data.memory = mem}; > >     switch (mem->model) { > @@ -3080,7 +3081,20 @@ qemuDomainAssignMemoryDeviceSlot(virDomainObj *vm, > >     case VIR_DOMAIN_MEMORY_MODEL_VIRTIO_PMEM: >     case VIR_DOMAIN_MEMORY_MODEL_VIRTIO_MEM: > -        return qemuDomainEnsurePCIAddress(vm, &dev); > +        if (mem->info.type == VIR_DOMAIN_DEVICE_ADDRESS_TYPE_NONE) { > +            if (qemuDomainIsS390CCW(vm->def)) > +                mem->info.type = VIR_DOMAIN_DEVICE_ADDRESS_TYPE_CCW; > +        } > + > +        if (mem->info.type == VIR_DOMAIN_DEVICE_ADDRESS_TYPE_NONE || > +            mem->info.type == VIR_DOMAIN_DEVICE_ADDRESS_TYPE_PCI) { > +            return qemuDomainEnsurePCIAddress(vm, &dev); > +        } else if (mem->info.type == > VIR_DOMAIN_DEVICE_ADDRESS_TYPE_CCW) { > +            if (!(ccwaddrs = > virDomainCCWAddressSetCreateFromDomain(vm->def))) > +                return -1; > +            return virDomainCCWAddressAssign(&mem->info, ccwaddrs, > +                                             !mem- > >info.addr.ccw.assigned); > +        } You leak ccwaddrs here, I'd suggest squashing something like this in: