Steps to reproduce the behavior:

1. Starting from a os-net-config configuration with drivers_autoprobe set to false:

   [cloud-admin@compute-0 ~]$ cat '/sys/devices/pci0000:80/0000:80:02.0/0000:82:00.3/sriov_drivers_autoprobe' 
   0
   

2. For a target NIC (e.g. nic6) create this os-net-configuration to enable drivers autoproble:

   [cloud-admin@compute-0 ~]$ cat /tmp/config.yaml 
   ---
   network_config:
   - type: sriov_pf
     name: nic6
     mtu: 9000
     numvfs: 5
     use_dhcp: false
     defroute: false
     nm_controlled: true
     hotplug: true
     promisc: false
     drivers_autoprobe: true
   
   - type: ovs_user_bridge
     name: br-dpdk1
     use_dhcp: false
     mtu: 9000
     members:
       - type: ovs_dpdk_port
         rx_queue: 1
         name: dpdk3
         members:
         - type: sriov_vf
           device: nic6
           vfid: 1
   

3. The apply the above config but using --noop option (which in theory Return the configuration commands, without applying them):

   os-net-config --noop -d -p nmstate -c /tmp/config.yaml 2>/dev/null
   ...
   2025-08-06 04:49:08.243 INFO os_net_config.utils._ordered_nics Active nics: ['eno1', 'eno2', 'enp4s0f0np0', 'enp6s0f0', 'enp6s0f1', 'enp6s0f2', 'enp6s0f3', 'enp130s0f0', 'enp130s0f1', 'enp130s0f2', 'enp130s0f3']
   2025-08-06 04:49:08.243 INFO os_net_config.common.get_pci_address enp130s0f3: pci address is 0000:82:00.3
   2025-08-06 04:49:08.243 INFO os_net_config.impl_nmstate.add_sriov_pf enp130s0f3: adding sriov pf
   2025-08-06 04:49:08.249 INFO os_net_config.common.get_pci_device_driver 0000:82:00.3: Fetching the PCI device driver
   2025-08-06 04:49:08.249 INFO os_net_config.utils.get_totalvfs enp130s0f3: sriov_totalvfs=32
   2025-08-06 04:49:08.251 DEBUG os_net_config.impl_nmstate.__dump_config ----------------------------
   2025-08-06 04:49:08.251 DEBUG os_net_config.impl_nmstate.__dump_config enp130s0f3: Prepared config
   accept-all-mac-addresses: true
   ethernet:
     sr-iov:
       drivers-autoprobe: true
       total-vfs: 5
   ethtool:
     feature:
       hw-tc-offload: false
   ipv4:
     auto-gateway: false
     dhcp: false
     enabled: false
   ipv6:
     auto-gateway: false
     autoconf: false
     dhcp: false
     enabled: false
   mtu: 9000
   name: enp130s0f3
   state: up
   type: ethernet
   
   2025-08-06 04:49:08.251 INFO os_net_config.common.get_pci_address enp130s0f3: pci address is 0000:82:0e.1
   2025-08-06 04:49:08.257 INFO os_net_config.common.get_default_vf_driver enp130s0f3-1: default vf driver is iavf
   2025-08-06 04:49:08.257 INFO os_net_config.common.get_pci_device_driver 0000:82:0e.1: Fetching the PCI device driver
   2025-08-06 04:49:08.257 INFO os_net_config.common.get_pci_device_driver 0000:82:0e.1: Fetching the PCI device driver
   2025-08-06 04:49:08.257 INFO os_net_config.common.set_driverctl_override 0000:82:0e.1: Binding with iavf
   driverctl --nosave set-override 0000:82:0e.1 iavf
   

4. Then the system hangs trying to apply driverctl --nosave set-override 0000:82:0e.1 iavf

   [cloud-admin@compute-0 ~]$ ps aux |grep driverctl
   root        8177  0.0  0.0   7256  3520 pts/1    S+   04:49   0:00 /usr/bin/bash /sbin/driverctl --nosave set-override 0000:82:0e.1 iavf
   cloud-a+    8219  0.0  0.0   6408  1920 pts/0    S+   05:04   0:00 grep --color=auto driverctl
   

Expected behavior

• No system commands must be applied with --noop