Andrea Riciputi wrote:
[]
[]
g++ -g -O2 -mpowerpc -no-cpp-precomp -mcpu=7450 -o .libs/t-locale
t-locale.o -L/sw/lib ../../tests/.libs/libtests.a
../../.libs/libgmpxx.dylib /sw/src/fink.bui
ld/gmp-4.1.4-13/gmp-4.1.4/.libs/libgmp.dylib -L/usr/lib/gcc/darwin/3.3
-L/usr/lib/gcc/darwin -L/usr/libexec/gcc/darwin/ppc/3.3/../../..
-lstdc++ -lgcc -lSystem
../../.libs/libgmp.dylib
ld: warning multiple definitions of symbol _localeconv
t-locale.o definition of _localeconv in section (__TEXT,__text)
/usr/lib/libSystem.dylib(localeconv.So) definition of _localeconv
creating t-locale
[snip]
make check-TESTS
PASS: t-cast
PASS: t-headers
PASS: t-ostream
operator>> input error
point ,
str ,5
FAIL: t-locale
What happens is that the t-locale test tries to redefine the
localeconv() function. It even has a test for checking whether this
redefinition was successful. Only this test in the test is lying:
It tests whether it can replace the use of localeconv() in libgmp.dylib,
which is a flat_namespace image. This works. But then it tries to use
the modified localeconv() with things from libgmpxx.dylib, which is a
twolevel_namespace image. This fails, because the library remembers that
localeconv comes from libSystem.
If you replace in the above compiler line the dynamic libgmpxx.dylib by
the static libgmpxx.a, it will work.
So it is the code of the test that is to blame, not the compiler.
OTOH I don't know why one of the dylibs is flat and the other one
twolevel...
--
Martin
-------------------------------------------------------
SF.Net email is sponsored by:
Tame your development challenges with Apache's Geronimo App Server. Download
it for free - -and be entered to win a 42" plasma tv or your very own
Sony(tm)PSP. Click here to play: http://sourceforge.net/geronimo.php
_______________________________________________
Fink-devel mailing list
Fink-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/fink-devel
