On Wed, Apr 07, 2004 at 01:54:40PM +0100, Steve Woodford wrote:
> On Wednesday 07 April 2004 1:33 pm, Ignatios Souvatzis wrote:
> 
> > On Wed, Apr 07, 2004 at 09:41:01AM +0300, Jukka Andberg wrote:
> > > - additionally, the method it uses to check the spl level
> > > (comparing return values from spl* functions) doesn't work very
> > > well on amiga as the m68k spl* functions seem set only the lower 16
> > > bits of the return value, the upper bits being 'random'.
> >
> > Uh-oh. and splfoo(0 is supposed to return int? Or do we have a spl_t
> > definition that we don't use yet?
> 
> This behaviour was something I introduced a while ago to speed up spl* 
> on m68k. It's quite normal and nothing to worry about.
> 
> Any, return values from the spl*() functions should be considered as 
> opaque cookies for use in a subsequent call to splx() only.

I completely agree. The check I added here was to try to catch some
missing spl() calls and was intended to be used on i386. This code shouldn't
be enabled for normal use anyway :)

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