1) The changes to the tr code in 2.1.107 truly do allow for the drivers to be totally independent, I know ibmtr still does some header manipulation, but olympic simply passes the whole packet up the protocol layers and it all works fine. The SysKonnect driver probably needs to be made into a simple packet reception / transmission driver rather that trying to do anything with the packet.
2) Header length: Yep, the maximum for this changed when the source routing went up from 18 to 30 bytes in the token ring spec.
3) TMS380 Chip: I'd be suprised if you can get a generic driver for this. The 380 is the chip that actually does all the clever electrical stuff to generate the signals on the wire. It is the actual card / bus interface chip that you'll need to get at. (e.g. the pci 3c359 I'm working on has the 380 on board, but the driver will be totally specific for this card, as I'm dealing mainly with pci / dma transfers and pci mmio commands.)
4) Multicast: Alan, I have a patch for this on the web site and Jochen has just sent me a patch to enable Option 2 multicasting in ibmtr. (I only did the main token ring code to get it to work with the olympic driver), I can put these together and send you a diff.
5) IPv6, Jochen has also recently supplied the IPv6 over token ring as well. I can also send this to you if interested (or Jochen can sent it direct, whatever works).
Mike
Linux Token Ring Project
http://www.linuxtr.net
-
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/