netdev - TAP programming and "br_xxx: received packet on br_xxx

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]

Message-Id: <629C1F6C-152A-4BFC-8A07-9F4E5B439325@redfish-solutions.com>
Date: Mon, 5 Aug 2024 12:42:20 -0600
From: Philip Prindeville <philipp_subx@...fish-solutions.com>
To: netdev@...r.kernel.org
Subject: TAP programming and "br_xxx: received packet on br_xxx_tap0 with own
 address as source address"

Hi,

I’m working on an L3 (non-IP) tunneling protocol that then emits the decapsulated payload with L2 wrappers.

I’ve got a TAP interface on a bridge with an Ethernet interface also bound to it [the bridge].

My question is, which address to source the L2 packets out I’m sending?  Do I use the Ethernet interface’s address, the bridge’s address, or the TAP interface’s address?

I’ve tried the latter two and in both cases I get some variation of:

br_xxx: received packet on br_xxx_tap0 with own address as source address (addr:xx:xx:xx:xx:xx, vlan:0)

What is the expected (correct) address to use in this paradigm?

Looking at br_fdb_update() it looks like I can’t use *any* local address, because they’ll all have BR_FDB_LOCAL set, won’t they?  Maybe I’m reading this wrong.

I’m using an embedded Linux distro that’s still at 5.15.19 so it would need to be applicable to that version.

Thanks,

-Philip