HEAD fails stress test on darwin

[Peter O'Gorman]

Hi,
This is just to remind me to bug Ralf when he gets back. I'd propose a 
patch, but I'm not quite sure of Ralf's intent with the test.

peter% ../../libtool/tests/testsuite -v 5
[snip]
stdout:
libtool: link: ar cru sub2/.libs/liba.a sub/.libs/a.o
libtool: link: ranlib sub2/.libs/liba.a
libtool: link: creating sub2/liba.la
libtool: link: ( cd "sub2/.libs" && rm -f "liba.la" && ln -s "../liba.la" 
"liba.la" )
../../libtool/tests/stresstest.at:170: $LIBTOOL --mode=link $CC $st -o 
"$rel"main "$rel"main.lo "$rel"sub2/liba.la
stdout:
libtool: link: gcc -o main .libs/main.o   sub2/.libs/liba.a
ld: multiple definitions of symbol _v1
.libs/main.o definition of _v1 in section (__DATA,__common)
sub2/.libs/liba.a(a.o) definition of _v1 in section (__DATA,__common)
../../libtool/tests/stresstest.at:171: ./main

Because both main.c and a.c have:
int v1;

Peter
--
Peter O'Gorman - http://www.pogma.com