Re: setsockopt() compat issue

[Manuel Bouyer <bouyer%antioche.eu.org@localhost>]

On Thu, Oct 16, 2008 at 07:30:14PM +0000, Christos Zoulas wrote:
> In article <20081016150444.GA1738%asim.lip6.fr@localhost>,
> Manuel Bouyer  <bouyer%antioche.eu.org@localhost> wrote:
> >On Thu, Oct 16, 2008 at 09:41:18AM -0500, Bryce Simonds wrote:
> >> Have you tried using a 32bit integer instead of a 64bit integer?
> >
> >No and it's not the point: this binary runs fine with a 3.1 kernel,
> >it should run with a -current one.
> 
> I am not sure if we should be providing backwards compatibility for buggy
> code. There were other related setsockopt changes that made the call return
> an error where before it would succeed. This is just one instance of them.
> If we really want to do this, we should create a compat_setsockopt call.

Well, maybe we should then. If we're claiming backward compatibility, we
should also be bug-compatible, or a software that did run, may not run any
more ...

-- 
Manuel Bouyer <bouyer%antioche.eu.org@localhost>
     NetBSD: 26 ans d'experience feront toujours la difference
--