> I don't think you are following the big picture of what I am saying. I
> was trying to follow Florian's intention (first make sure I understand
> it) and suggest that the FCS checking code in the patch he submitted
> is not doing what it was intended to. I am getting apparent FCS
> mismatches reported by the program, when I know full well that the MAC
> I am testing on would have dropped those frames were they really
> invalid.
I think this FCS check is not needed. If we feed the MAC random data,
something like 1 in 65535 will have a valid FCS and get passed up.
I've not seen this happen with Ethernet, but i have seen other network
technologies wrong decoding noise on the line and passing up frames
with around 1 in 65536 probability.
But then having the correct Ethertype is another 1 in 65536. So it
seem pretty improbably we do receiver a packet in this method which is
bad. So i would say, any packet received here is a good packet, and
indicate the RGMII mode works. If we don't receive a packet, the mode
is very probably bad.
Andrew
