At Sun, 9 Jul 2006 18:11:08 +0000 (UTC),
Christos Zoulas wrote:
> 
> In article <20060709165010.GB293@beta.martani.repy.czf>,
> Pavel Cahyna  <pavel@NetBSD.org> wrote:
> >On Sun, Jul 09, 2006 at 04:17:11PM +0200, Peter Postma wrote:
> >> This patch has been a while in my tree and I completely forgot it...
> >> 
> >> It moves the PF_GENERATED m_tag to the new packet in icmp_error.
> >> This is needed because the pf code can call icmp_error with setting
> >> this tag, but the new packet should not be filtered when it comes back
> >> to pf(4).
> >
> >To me, introducing pf-specific code outside dist/pf looks like an
> >abstraction violation.
> 
> PF_GENERATED means generated by "a" packet filter, not necessarily "pf".

Just a nit: we have been using "pfil" (for some years now) with the
meaning "packet filter".

I reacted the same way Pavel did, but since I don't suspect ipfilter
is going to use this the name is fine.