My planned work on networking stack

[Vladimir Grebenschikov]

On вт, 2004-03-02 at 19:32 +0000, Bruce M Simpson wrote:

> On Tue, Mar 02, 2004 at 07:09:02PM +0300, Gleb Smirnoff wrote:
> >   I do not insist that AS pathes in kernel are good idea. If you show me an
> > other way to get AS information when constructing netflow exports in kernel,
> > I'd be thankful. I'd be also thankful if you describe how policy routing can be
> > implemented while no AS info in kernel.
> >   What do other FreeBSD networking withards think?
> 
> I don't see any reason why we couldn't accept, for example, a 32-bit cookie
> for abuse by a userland daemon, with pid, as it pleases (via an rtmsg
> extension and PF_ROUTE). That is generic enough to provide the tie-in
> needed with the userland RIB and the kernel FIB.

This possible solution when you run accounting, but if you are going to
do policy-routing, you need to get routing decision based on additional
route information, and in this case you need keep this additional data
in kerenl. But I agree with you that rtentry is bad place for that
information.

> ABI breakage may occur, but I would consider that the PF_ROUTE code is in need
> of an overhaul anyway (see my mail to ru@ from some months ago on -current or
> -net with code able to panic a kernel through malformed rtmsg contents).

We need ability to link rtentry records with custom kernel data handled
by extension modules (such as cisco-like accounting or policy-routing)

> BMS