netdev - How to avoid getting ndo_open() immediately after probe

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-ID: <CO6PR18MB4465B4170C7A3B8F6DEFB369D4699@CO6PR18MB4465.namprd18.prod.outlook.com>
Date:   Thu, 2 Dec 2021 18:11:30 +0000
From:   Arijit De <arijitde@...vell.com>
To:     "netdev@...r.kernel.org" <netdev@...r.kernel.org>
Subject: How to avoid getting ndo_open() immediately after probe

Hi,

I have handled the probe() and registered the netdev structure using register_netdev().
I have observed in other opensource driver(i.e. Intel e1000e driver) that ndo_open() gets called only when we try to bring up the interface (i.e. ifconfig <ip> ifconfig eth0 <ip-addr> netmask <net-mask> up).
But in my network driver I am getting ndo_open() call immediately after I handle the probe(). It's a wrong behavior, also my network interface is getting to UP/RUNNING state(as shown below) even without any valid ip address.

enp1s0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        ether 00:22:55:33:22:28  txqueuelen 1000  (Ethernet)
        RX packets 0  bytes 0 (0.0 B)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 252  bytes 43066 (43.0 KB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions

What is the change required in the driver such that my network interface(enp1s0) should be in down state(BROADCAST & MULTICAST only) after probe()
and ndo_open() call should happen only when device gets configured?

Thanks
Arijit