[PATCH] linux: generic: net: dsa: mv88e6xxx LED support

[Tobias Waldekranz]

On sön, okt 06, 2024 at 21:50, Linus Walleij <linus.walleij at linaro.org> wrote:
> On Sun, Oct 6, 2024 at 6:23 PM Daniel Golle <daniel at makrotopia.org> wrote:
>
>> On Sun, Oct 06, 2024 at 05:32:04PM +0200, Linus Walleij wrote:
>> > This solution that is needed for some routers to provide proper
>> > LED activity when controlled directly by the MV88E6xxx switch,
>> > has just been merged in upstream Linux.
>> > [...]
>> > diff --git a/target/linux/generic/pending-6.6/714-net-dsa-mv88e6xxx-Support-LED-control.patch b/target/linux/generic/pending-6.6/714-net-dsa-mv88e6xxx-Support-LED-control.patch
>> > new file mode 100644
>> > index 000000000000..8557fe2510b8
>> > --- /dev/null
>> > +++ b/target/linux/generic/pending-6.6/714-net-dsa-mv88e6xxx-Support-LED-control.patch
>> > [...]
>> > diff --git a/target/linux/generic/pending-6.6/768-net-dsa-mv88e6xxx-Request-assisted-learning-on-CPU-port.patch b/target/linux/generic/pending-6.6/768-net-dsa-mv88e6xxx-Request-assisted-learning-on-CPU-port.patch
>> > index 3f3d7572e015..92f87f619873 100644
>> > --- a/target/linux/generic/pending-6.6/768-net-dsa-mv88e6xxx-Request-assisted-learning-on-CPU-port.patch
>> > +++ b/target/linux/generic/pending-6.6/768-net-dsa-mv88e6xxx-Request-assisted-learning-on-CPU-port.patch
>>
>> As both patches are now already part of mainline Linux they should go
>> into target/linux/generic/backport-6.6/ with a 'v6.13' prefix (as they
>> are expected to be part of Linux v6.13).
>
> I looked into Tobias' patch for assisted learning on the CPU port
> and it is actually not merged upstream, it was part of an RFC
> series in 2021 and I don't know what happened with that or
> why it is still part of the OpenWrt patch set, the rest of that RFC
> patch set is not in OpenWrt so I have no idea what the story is
> here...
> https://lore.kernel.org/netdev/20210116012515.3152-8-tobias@waldekranz.com/
>
> It came into OpenWrt here:
> https://git.openwrt.org/?p=openwrt/openwrt.git;a=commit;f=target/linux/generic/pending-5.4/768-net-dsa-mv88e6xxx-Request-assisted-learning-on-CPU-port.patch;hb=920eaab1d8179035d0ae1047e75cf9a50da6a6eb
>
> From context it seems like Tobias' patch is something that should
> have been merged upstream ages ago but fell between the cracks
> because it happened to be in an unrelated RFC series.
>
> Tobias is this something you/we should push upstream?

No, we should make sure that it is removed from OpenWRT's kernels :)

Since 041bd545e124, learning is enabled on the CPU port. In combination
with d82f8ab0d874 (tx forward offloading), i.e., we send FORWARD frames
from the CPU when forwarding packets from bridges to hardware, it means
that the switch now automatically picks up the "local" addresses of
interest.

In short: we have hardware CPU port learning, so there's no need for
any assistance.

I'm afraid my memory is too short to remember how it ended up in
OpenWRT.

--
Tobias