> For the record, I think that this sort of warning from gcc is completely 
> bogus.  If we want meaningful warnings, we should turn on those which 
> produce meaningful warnings ... -Wall often produces noise.

I consider this one to be a useful one--it's saved me on a couple of
occasions when making changes to someone else's code.

> ...y'know, it strikes me as interesting that this warning doesn't show up 
> in a SPARC kernel build... "weird" :-)

I really think it's a bug in gcc (see my other message or look at the
code in question to see that this is a case that gcc should be able to
catch properly).  It could be in the m68k md code, but I'm not sure how.

-allen

-- 
  Allen Briggs - end killing - briggs@bev.net ** MacBSD == NetBSD/mac68k **
   Where does all my time go?  <a href="http://www.netbsd.org/">Guess.</a>