hw/pci-host/pnv_phb: Avoid quitting QEMU if hotplug of pnv-phb-root-port fails

Currently QEMU terminates if you try to hotplug pnv-phb-root-port in
an environment where it is not supported, e.g. if doing this:

 echo "device_add pnv-phb-root-port" | \
 ./qemu-system-ppc64 -monitor stdio -M powernv9

To avoid this problem, the pnv_phb_root_port_realize() function should
not use error_fatal when trying to set the properties which might not
be available.

Fixes: c2f3f78af5 ("ppc/pnv: set root port chassis and slot using Bus properties")
Signed-off-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Cédric Le Goater <clg@kaod.org>
Reviewed-by: Daniel Henrique Barboza <danielhb413@gmail.com>
Message-Id: <20221109122210.115667-1-thuth@redhat.com>
Signed-off-by: Daniel Henrique Barboza <danielhb413@gmail.com>

diff --git a/hw/pci-host/pnv_phb.c b/hw/pci-host/pnv_phb.c
index 7b11f1e8ddf6c2d50e8787aa0a426377e3e21618..0b26b43736f2a5725e1d9041623ca660cae292a7 100644 (file)
--- a/hw/pci-host/pnv_phb.c
+++ b/hw/pci-host/pnv_phb.c
@@ -241,8 +241,16 @@ static void pnv_phb_root_port_realize(DeviceState *dev, Error **errp)
      * QOM id. 'chip_id' is going to be used as PCIE chassis for the
      * root port.
      */
-    chip_id = object_property_get_int(OBJECT(bus), "chip-id", &error_fatal);
-    index = object_property_get_int(OBJECT(bus), "phb-id", &error_fatal);
+    chip_id = object_property_get_int(OBJECT(bus), "chip-id", &local_err);
+    if (local_err) {
+        error_propagate(errp, local_err);
+        return;
+    }
+    index = object_property_get_int(OBJECT(bus), "phb-id", &local_err);
+    if (local_err) {
+        error_propagate(errp, local_err);
+        return;
+    }
 
     /* Set unique chassis/slot values for the root port */
     qdev_prop_set_uint8(dev, "chassis", chip_id);