RE: Token Ring Bug (Broadcast and Multicast address check)

Tom Gallagher (Tom.Gallagher@Madge.com)
Thu, 6 May 1999 17:23:10 +0100


>
> Hello!
>
> > byte with 0x80 and not 1 when checking for a broadcast or multicast
> > address.
>
> Excellent! This puzzle bedevils me for year or so...
> And nobody was able to give any intelligible answer 8)
>
> Let me one question to complete the picture: the same bit
> of tr source address is TR_RII. Is it true? Do another magic
> bits exist?

Yes this is true. Source addresses are always individual addresses so
the individual/group bit used in the destination address is instead
used to indicate source routing information in the source address.
Other magic bits do exist:-
In a destination address bit 6 (mask 0x40) of the first byte indicates
a locally administered address (if 1).
Also if both bit 7 and bit 6 are set (destination address starts with 0xC0)
then bit 7 of the second address byte indicates whether the address is
a group (1) or functional address (0).

> Also, could you write tr_header_parse routine to fetch
> source address? (Look at eth_header_parse)
>
> BTW, could you send me piece of "tcpdump -p -e" with samples
> of multicast packets originated by another hosts?

I'll look into this when I get time.

Thanks for the response.

from Tom

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/