netdev - Re: [RFC] net: dsa: mv88e6xxx disable IGMP snooping on cpu port

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>] [day] [month] [year] [list]

Date:   Mon, 27 Mar 2023 17:26:10 +0200
From:   Andrew Lunn <andrew@...n.ch>
To:     Steffen Bätz - Innosonix GmbH 
        <steffen@...osonix.de>
Cc:     Fabio Estevam <festevam@...il.com>, olteanv@...il.com,
        davem@...emloft.net, kuba@...nel.org, netdev@...r.kernel.org,
        Fabio Estevam <festevam@...x.de>
Subject: Re: [RFC] net: dsa: mv88e6xxx disable IGMP snooping on cpu port

On Mon, Mar 27, 2023 at 04:51:20PM +0200, Steffen Bätz - Innosonix GmbH wrote:
> Hi Andrew,
> Yes, strangely, this piece of code has stayed the same for years.
> But we only face this behaviour if you add a bridge interface on top of the DSA
> ports.
> The setup looks like: eth0 (physical port of the imx8mn) <-> lan3/lan4 (DSA
> ports of the mv88e6320)
> 
> ip link add br0 type bridge
> ip link set br0 up
> ip link set lan3 master br0
> ip link set lan4 master br0
> ip link set lan3 up
> ip link set lan4 up
> dhcpcd -b br0
> 
> If now you try to receive an audio multicast stream, like from 239.255.84.1,
> the neighbour bridge will not forward this stream to our board. And there is no
> entry in the MDB of the external switch.

Ah, O.K. That is not actually IGMP snooping. It is normal usage of
IGMP. I probably did not test that. I think all my multicast source
and sinks where on switch ports, and i tested that the switch did not
flood multicast to all ports, just ports with listeners.

So your change looks O.K, but as i said, it probably should apply to
DSA ports as well as CPU ports. And i suggest you reword the commit
message to differentiate between IGMP snooping and an IGMP listener on
the bridge.

    Andrew