Re: Another endianess problem on Solaris

[Martin Kahlert <martin dot kahlert at infineon dot com>]

Hi Jeff,
thanks for your reply.

On Fri, Feb 08, 2002 at 09:52:21AM -0500, Jeff Sturm wrote:
> On Fri, 8 Feb 2002, Martin Kahlert wrote:
> > I think, i found an endianess problem on Solaris with boolean:
> 
> Yes, it's broken.  This has been discussed at length.  We had a working
> patch once and didn't commit it; I was going to revamp the return types in
> libffi first.  I dropped it there.  No time for hacking now :(
> 
> Strictly speaking it's not a libffi problem.  (I'm looking for the
> archives, but htdig doesn't seem to be up.)

I guess i found it at
http://gcc.gnu.org/ml/java-patches/2001-q4/msg00236.html
Your patch seems to apply to today's natMethod.cc without too much trouble.
I will try it on my Solaris build now.

It would be nice, if there was a solution found for 3.1. At least, there
should be an entry in the 'known bugs' section. It took me quite a while to
find the problem.

Is Solaris the only gcj-port, where these problems exist?
Is Linux-Alpha big endian, for example?

Thanks alot,
Martin.

-- 
The early bird catches the worm. If you want something else for       
breakfast, get up later.