Re: Dealing with M_HASFCS for protocols that do not do ethernet crc

[Robert Swindells <rjs%fdy2.co.uk@localhost>]

Martin Husemann <martin%duskware.de@localhost> wrote:
>On Tue, Aug 09, 2022 at 09:17:59PM +0700, Robert Elz wrote:
>> The relevant mbuf bit, as I remember it, simply allows the driver to
>> inform the stack that the FCS is still there in the frame, and hasn't
>> been removed, either by the hardware, or by the driver.   That I believe
>> is largely what Joerg suggested it means.
>
>But that means somewhere in the stack we have a bug biting Chris'
>use of ethertalk where some code not properly deals with M_HASFCS.

I can't see how this would happen, ether_input() strips the FCS if
M_HASFCS is set for the mbuf.

>This also means the commit should be backed out, or am I misunderstanding
>again?

I think it should.

We could also try to enable FCS stripping for drivers that currently
don't do it, doing it for eqos(4) looks easy enough, if someone has the
documentation for dwc_gmac could they lookup if it is possible to do
this?