netdev - Re: [PATCH net] bonding: fix mii

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

Message-ID: <fbc92957-4cf4-4687-bc2d-ed09cedf8572@lunn.ch>
Date: Fri, 7 Nov 2025 14:55:24 +0100
From: Andrew Lunn <andrew@...n.ch>
To: Nicolas Dichtel <nicolas.dichtel@...nd.com>
Cc: "David S . Miller" <davem@...emloft.net>,
	Jakub Kicinski <kuba@...nel.org>, Paolo Abeni <pabeni@...hat.com>,
	Eric Dumazet <edumazet@...gle.com>, Jay Vosburgh <jv@...sburgh.net>,
	Andrew Lunn <andrew+netdev@...n.ch>,
	Stanislav Fomichev <sdf@...ichev.me>, netdev@...r.kernel.org
Subject: Re: [PATCH net] bonding: fix mii_status when slave is down

On Fri, Nov 07, 2025 at 09:10:58AM +0100, Nicolas Dichtel wrote:
> Le 07/11/2025 à 03:36, Andrew Lunn a écrit :
> > On Thu, Nov 06, 2025 at 07:02:52PM +0100, Nicolas Dichtel wrote:
> >> netif_carrier_ok() doesn't check if the slave is up. Before the below
> >> commit, netif_running() was also checked.
> > 
> > I assume you have a device which is reporting carrier, despite being
> > admin down? That is pretty unusual, and suggests its PHY handing is
> > broken. What device is it? You might want to also fix it.
> Yes, one slave is put down administratively. Before the mentioned commit, the
> status was correctly reported; it's no longer the case.
> It's a regression.

I agree with your fix, but i would also like to know more about the
interfaces you are testing on. We should probably fix that device as
well. What is it?

Things do get 'interesting' when there is a BMC also sharing the
link. Linux cannot tell the PHY to drop the carrier because that would
also disconnect the BMC, but there should be some separation between
netif_carrier_ok() returns and the state of the PHY.

	Andrew