[net-next,v2,0/3] net: dsa: mv88e6xxx: fix IPv6

We have had some emails in private over this issue, this is my current
patch set rebased on top of net-next which provides working IPv6 (and
probably other protocols as well) over mv88e6xxx DSA switches.

The problem comes down to mv88e6xxx defaulting to not flood unknown
unicast and multicast datagrams, as they would be by dumb switches,
and as the Linux bridge code does by default.

These flood settings can be disabled via the Linux bridge code if it's
desired to make the switch behave more like a managed switch, eg, by
enabling the multicast querier.  However, the multicast querier
defaults to being disabled which effectively means that by default,
mv88e6xxx switches block all multicast traffic.  This is at odds with
the Linux bridge documentation, and the defaults that the Linux bridge
code adopts.

So, this patch set adds DSA support for Linux bridge flags, adds
mv88e6xxx support for the unicast and multicast flooding flags, and
lastly enables flooding of these frames by default to match the
Linux bridge defaults.

 drivers/net/dsa/mv88e6xxx/chip.c | 45 ++++++++++++++++++++++++++++++++++------
 include/net/dsa.h                |  3 +++
 net/dsa/dsa_priv.h               |  2 ++
 net/dsa/port.c                   | 15 ++++++++++++++
 net/dsa/slave.c                  |  6 ++++++
 5 files changed, 65 insertions(+), 6 deletions(-)

v2: fix a couple of compile errors in patch 2 and patch 3 (oops).

On Sun, Feb 17, 2019 at 04:31:14PM +0000, Russell King - ARM Linux admin wrote:
> We have had some emails in private over this issue, this is my current
> patch set rebased on top of net-next which provides working IPv6 (and
> probably other protocols as well) over mv88e6xxx DSA switches.
> 
> The problem comes down to mv88e6xxx defaulting to not flood unknown
> unicast and multicast datagrams, as they would be by dumb switches,
> and as the Linux bridge code does by default.
> 
> These flood settings can be disabled via the Linux bridge code if it's
> desired to make the switch behave more like a managed switch, eg, by
> enabling the multicast querier.  However, the multicast querier
> defaults to being disabled which effectively means that by default,
> mv88e6xxx switches block all multicast traffic.  This is at odds with
> the Linux bridge documentation, and the defaults that the Linux bridge
> code adopts.
> 
> So, this patch set adds DSA support for Linux bridge flags, adds
> mv88e6xxx support for the unicast and multicast flooding flags, and
> lastly enables flooding of these frames by default to match the
> Linux bridge defaults.

While looking at some of the other DSA drivers, I've noticed that
others are also programmed to forward unknown frames to the CPU
port.  Does this not end up breaking stuff?

If I tcpdump the ethernet interface for the CPU port, what I see
is:

11:21:21.901127 00:22:68:15:37:dd (oui Unknown) > 52:54:00:00:06:25 (oui
Unknown), ethertype MEDSA (0xdada), length 126: Forward, untagged,
dev.port:vlan 0.4:0, pri 0: ethertype IPv6 (0x86dd)
e0022681537dd.dyn.armlinux.org.uk > tftp.armlinux.org.uk: ICMP6, echo
request, seq 1, length 64

which is the unknown frame being delivered to the CPU port.  It seems
nothing else happens with the frame - it is ignored.  Before my fixes
for mv88e6xxx, that frame (and the following frames for the same MAC
address) would end up being forwarded only to the CPU port and dropped
on the floor, never making their way to their intended destination.

It seems that "the hardware doesn't know what to do, forward it to
Linux to sort out" doesn't actually work.