Re: [PATCH-1.5] Clean up darwin stuff

[Peter O'Gorman]

Gary V. Vaughan wrote:

> Peter O'Gorman wrote:
>
> > Okay to apply?
>
>
> The vast majority of this patch seems to bring branch-1-5 in line with
> HEAD, which is fine.  However, I can't find the analog of this chunk:
>
>
> > @@ -5404,7 +5367,7 @@ $echo "local: *; };" >> $output_objdir/$
> >       ;;
> >
> >     darwin* | rhapsody*)
> > -    if test "$GXX" = yes ; then
> > +    if test "$GCC" = yes ; then
> >       _LT_AC_TAGVAR(archive_cmds_need_lc, $1)=no
> >       case "$host_os" in
> >       rhapsody* | darwin1.[[012]])
>
>
> Around line 3800 of m4/libtool.m4 in HEAD is a similar chunk, but the
> GCC vs GXX test is done very differently.
>
> What does this chunk fix?  If the testsuite needs this chunk to pass, then
> go ahead and commit the whole lot.  If the testsuite passes with just the
> rest of the patch (minus the chunk above), please commit all but the chunk
> above anyway while we nail this down.

The error was in my first attempt to fix libtool on darwin sometime last 
year, it should always have been testing for GCC not GXX at this point. The 
reason to test is that non gcc compilers do not work.

I shall now proceed to my next trick, to backport the xlc stuff from head to 
branch-1-5. Maybe... if I have time :)

Committed patch as is,
Thanks,
Peter
--
Peter O'Gorman - http://www.pogma.com