Re: git: 7e51bc6cdd5c - main - pf: Introduce unhandled_af()

On 5 Mar 2025, at 17:43, Gleb Smirnoff wrote:
> On Tue, Mar 04, 2025 at 08:05:51AM +0000, Kristof Provost wrote:
> K> The branch main has been updated by kp:
> K>
> K> URL: https://cgit.FreeBSD.org/src/commit/?id=7e51bc6cdd5c317109e25b0b64230d00d68dceb3
> K>
> K> commit 7e51bc6cdd5c317109e25b0b64230d00d68dceb3
> K> Author:     Kristof Provost <kp@FreeBSD.org>
> K> AuthorDate: 2025-03-03 16:26:39 +0000
> K> Commit:     Kristof Provost <kp@FreeBSD.org>
> K> CommitDate: 2025-03-04 08:05:37 +0000
> K>
> K>     pf: Introduce unhandled_af()
> K>
> K>     For cases where code conditionally does something based on an address family
> K>     and later assumes one of the paths was taken.  This was initially just calls
> K>     to panic until guenther suggested a function to reduce the amount of strings
> K>     needed.
> K>
> K>     This reduces the amount of noise with static analysers and acts as a sanity
> K>     check.
> K>
> K>     ok guenther@ bluhm@
> K>
> K>     Obtained from:  OpenBSD, jsg <jsg@openbsd.org>, ba4138390b
> K>     Sponsored by:   Rubicon Communications, LLC ("Netgate")
>
> Heh, enum solves the problem at compilation time.  I will try to revisit the review
> that had this idea.
>
Oh yeah, I’d forgotten about that one.

In any event, this is useful as a diff-reduction exercise, if nothing else. I’m hoping to be able to continue importing OpenBSD patches, but there’s still a very long way to go before we’re all the way caught up.

And even if/when we get there there’ll still be a substantial delta, because we don’t want to drop support for older syntax in the way OpenBSD has, and we want to keep stuff like your multicore work and the vnet support. And the SCTP support. And the basic Ethernet support.

Anyway, lots of random thoughts to say: that’d be nice to have, but certainly isn’t urgent.

Best regards,
Kristof