DragonFly kernel List (threaded) for 2003-11
Re: Am I way off base here?
:Matthew Dillon wrote:
:> Is there something fundamentally difficult about implementing
:> things like getpwnam() through an IPC mechanism + maintainng a local
:> cache that I am missing? Because I am really getting angry at the
:> FreeBSD-5 folk acting so goddamn high and mighty about their NSS DLL
:> requiring a dynamic root crap.
:The issue seems to be that NSS must be supported immediately and
:there's no current implementation your IPC NSS mechanism.
:> NIS never implemented it right. What am I missing here?
:The answer is to implement it properly, then share the code. I'm
:losing faith in both sides at this point.
:The behaviour of the FreeBSD folks is par for the course.
:Whenever someone proposes an idea that conflicts with the path
:they've already taken, they pout. The better the idea and the
:more foolish it makes their chosen path look, the higher they
:raise their bottom lip. And now there's apparently a
:new rule, whenever someone mentions DragonFly they cover their
:ears and scream loudly to make sure they can't hear it.
:But still, the answer is sharing better code implementing the
:better idea. And if you're still angry at that point, you've
:earned the right to say "I told you so."
Well, my behavior isn't much better when it gets into the flame domain :-)
In anycase, the two implementations are not diametrically opposed but
they are implemented differently. My argument isn't really with the
design choice, though I do feel that the IPC method is better. My
argument is with FreeBSD making that design choice but then not following
through with the necessary static binary support to really make it work
well, and then using that as an excuse to make root dynamic. Sigh.