Jeff Garzik <garzik@havoc.gtf.org> writes:
> "SIOCDEVICE" as a constant is unacceptable, so it would need to be
> SIOCWANDEVICE or something similar.
Well, I was probably under impression it should be used for Ethernet
as well (see the Dec 2000 thread)... Now I think I know people
using Ethernet (with full duplex over SM fibre) for WAN connections
- so SIOCWANDEVICE is ok. Not sure about TR, though - anyone using
it for WAN networking?
> SIOCETHTOOL, for example, is an ioctl which actually provides
> sub-ioctls, so that is probably a good model to follow.
SIOCDEVICE^WSIOCWANDEVICE of course has sub-ioctls as well. It is
obviously impossible without them.
I do _not_ want to fight any ETHTOOL vs SIOCDEVICE etc. battle here.
What I want is creating the best interface for controlling network
devices. Including Token Ring and Ethernet, unless there are valid
reasons to do otherwise.
I think we should concentrate on the interface first, then I will
patch the HDLC implementation.
If we're here... maybe we should really drop using the ifreq structure
and _replace_ it with better one (variable-sized)? It can be done
gradually, as both are quite compatible.
-- Krzysztof Halasa Network Administrator - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Thu Feb 07 2002 - 21:00:33 EST