Sandy Harris wrote:
> jamal wrote:
> > > What problem does this fix?
> > >
> > > If you are mucking with the ifindex, you may be affecting many places
> > > in the rest of the kernel, as well as user-space programs which use
> > > ifindex to bind to raw devices.
> > I am talking about 2.5 possibilities now that 2.4 is out. I think
> > "parasitic/virtual" interfaces is not a issue specific to VLANs.
> > VLANs happen to use devices today to solve the problem.
> > As pointed by that example no routing daemons are doing aliased
> > interfaces (which are also virtual interfaces).
> > We need some more general solution.
> Something like this also becomes an issue when you want routing
> daemons to interact sensibly with IPSEC tunnels. A paper on these
> issues is at:
> It is not (AFAIK) clear that the FreeS/WAN team will adopt the solutions
> suggested there, but it is very clear we need to deal with those issues.
Hrm, what if they just made each IP-SEC interface a net_device? If they
are a routable entity, with it's own IP address, it starts to look a lot
like an interface/net_device.
This has seeming worked well for VLANs: Maybe net_device is already
So, what would be the down-side of having VLANs and other virtual interfaces
be net_devices? The only thing I ever thought of was the linear lookups,
which is why I wrote the hash code. The beauty of working with existing
user-space tools should not be over-looked!
It may be easier to fix other problems with many interface/net_devices
than cram a whole other virtual net_device structure (with many duplicate
functionalities found in the current net_device).
Ben Greear (greearb@xxxxxxxxxxxxxxx) http://www.candelatech.com
Author of ScryMUD: scry.wanfear.com 4444 (Released under GPL)