Re: fibnum2.diff (Re: per-FIB socket binding)

On Thursday, February 20, 2025 4:47:41 PM UTC Mark Johnston wrote:
> On Tue, Feb 18, 2025 at 05:16:07AM +0000, Paul Vixie wrote:
> > this is the second fibnum patch, ...
> 
> The high-level changes seem to be:
> - If a TCP listening socket's FIB is 0, then the FIB of incoming
>   connections matched to that socket will be the FIB of the interface on
>   which the initial SYN was received.
> - When a socket is bound to a specific address, its FIB is set to the
>   FIB of the interface which owns the address.

yes.

> The logic seems reasonable to me.
> 
> The handling of the second point seems incomplete: it doesn't update the
> FIB number stored in the socket itself.  Gleb and I talked a bit about
> eliminating that field entirely so that there's only one source of
> truth, and I think we'll eventually do that, but in the meantime,
> in_pcbbind() needs to update so_fibnum as well.

the real fix is to not have an architecture that requires things like this in 
in_pcballoc(), which triggered a Hairball Alert during the fibnum2 work:

>         inp->inp_pcbinfo = pcbinfo;
>         inp->inp_socket = so;
>         inp->inp_cred = crhold(so->so_cred);
>         inp->inp_inc.inc_fibnum = so->so_fibnum;

i'm totally willing to floss all of this out and get us down to a single 
source of truth if there are willing reviewers for it. meanwhile, i'll add the 
corresponding logic to in_pcbbind(), with an XXX marker.

> > comments, questions, and especially testing results would be very welcome.
> 
> This patch doesn't apply to main.  Did you write it against stable/14?

i've been working in a 14.2 context but i'll foreport and test.

thanks for engaging.

-- 
Paul Vixie