I don't have this hardware but it looks like we weren't adding bridge
devices as intended.  Maybe the bridge is always the last device?
Fixes: 05b125004815 ("PCI: cpcihp: Iterate over all devices in slot, not functions 0-7")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Acked-by: Yijing Wang <wangyijing@huawei.com>
CC: stable@vger.kernel.org	# v3.9+
        }
        parent = slot->dev->bus;
 
-       list_for_each_entry(dev, &parent->devices, bus_list)
+       list_for_each_entry(dev, &parent->devices, bus_list) {
                if (PCI_SLOT(dev->devfn) != PCI_SLOT(slot->devfn))
                        continue;
                if (pci_is_bridge(dev))
                        pci_hp_add_bridge(dev);
+       }
 
 
        pci_assign_unassigned_bridge_resources(parent->self);