Re: [PATCH v8 net-next 2/2] net: dsa: mv88e6xxx: mac-auth/MAB implementation

[netdev]

On 2022-11-20 16:00, Vladimir Oltean wrote:
> On Sun, Nov 20, 2022 at 11:21:08AM +0100, netdev@xxxxxxxxxxxxxxxxxxxx 
> wrote:
> > I have something like this, using 'mvls vtu' from
> > https://github.com/wkz/mdio-tools:
> >  VID   FID  SID  P  Q  F  0  1  2  3  4  5  6  7  8  9  a
> >    0     0    0  y  -  -  =  =  =  =  =  =  =  =  =  =  =
> >    1     2    0  -  -  -  u  u  u  u  u  u  u  u  u  u  =
> > 4095     1    0  -  -  -  =  =  =  =  =  =  =  =  =  =  =
> >
> > as a vtu table. I don't remember exactly the consequences, but I am 
> > quite
> > sure that fid=0 gave
> > incorrect handling, but there might be something that I have missed as 
> > to
> > other setups.
>
> Can you please find out? There needs to be an answer as to why 
> something
> which shouldn't happen happens.

Just an update on this, as when running the selftests now, I have 
experienced
the case where fid=0 in the interrupt handler. The reported mac is the 
same
as the one where the handling was successful in the selftest.

So I don't know what causes this fid=0 event, maybe some timing with the 
chip
op when stressed resulting in erroneous reading of fid?