[PING][Patch]: multilib fixes for libjava

Jack Howarth howarth@bromo.msbb.uc.edu
Fri Nov 17 23:29:00 GMT 2006

    Can we move the multilib fixes into gcc 4.2 branch
as well after a few weeks of testing in gcc trunk? I tend
to view this as fixing a regression in gcc 4.2 compared
to 4.1. Without these changes, one has to pass a new 
--disable-multilib option to configure when building
the fortran or java targets on Darwin machines with
32-bit processors. I would argue that that amounts to
a regression from gcc 4.1.1.

On Thu, Nov 16, 2006 at 11:31:40PM -0800, Mike Stump wrote:
> On Nov 16, 2006, at 7:51 AM, Andrew Haley wrote:
> >Lest anyone think this is waiting on me, let me say that I have no
> >problem with this.  It is for Darwin maintainers to comment.
> Didn't you see: http://gcc.gnu.org/ml/gcc-patches/2006-11/ 
> msg00518.html in the thread.  :-)
> The patch is approved, however, do keep an eye out for other  
> multilibs that wonder by that might trip over it.  Or put another  
> way, don't check it in and then take a 2 week vacation.  :-)
> >But they won't see it, because Darwin does not appear in the Subject
> >line and libjava does.
> I see it all (as long as it hits gcc-patches or gcc).

More information about the Gcc-patches mailing list