Ivan Passos wrote:
> Imagine a card that implements PPP in its own firmware (i.e. it doesn't
> use pppd), and this firmware talks to a driver in the kernel. How can this
> card do user authentication??
IMO, by far the best way to do it (at least in 2.4) would be to have the
hardware send and receive PPP packets, and then write a new handler on
top of ppp_generic and make pppd talk to that. That way the CPU-intensive
stuff (like escaping characters and such) would be handled by your hardware,
and pppd would still be used for authentication and such. This has the
big advantage that the user doesn't have to change their setup much to
make it work.
A similar option would be to present a tty driver that somehow hijacks the
PPP line discipline and handles it in hardware. Not sure how easy that would
be to do, but that would even work on older kernels.
If the hardware really, really, REALLY can't do this and must implement the
full ppp state machine in hardware (with negligable performance advantage),
then I'd still suggest doing it as above. Only now you'll need to exchange
fake LCP/PAP/IPCP packets with pppd inorder to let it handle authentication
and such.
-Mitch
-
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/
This archive was generated by hypermail 2b29 : Mon Jul 31 2000 - 21:00:28 EST