christian pellegrin wrote: > On Mon, Feb 1, 2010 at 7:01 PM, Paul Thomas <[email protected]> wrote: > >> Yes, I had originally thought osm was needed because when I watched an open >> bus.
Ah! >> I would see continuous traffic after a send. I didn't want to get into >> a situation where if one of 4 nodes died the retries to the dead node ate up >> the whole bus, but that's NOT what I see. As long as there is 1 good node, a >> packet to a non-existent node is not retried. Indeed. And as commonly agreed on the ML the CAN controller tries to send the frame until it gets ACKed or we reach the bus-off condition. There's no need for any OSM :-) > So ASAP I will prepare the patch for netdev-next without OSM for the mcp2515. Yes. That is a good idea. Regards, Oliver _______________________________________________ Socketcan-core mailing list [email protected] https://lists.berlios.de/mailman/listinfo/socketcan-core
