Daniel Macks <dma...@netspace.org> said: > On Fri, Jun 11, 2010 at 06:57:50AM -0400, Alexander Hansen wrote: >> Yup, 10.6/32-bit. 10.6/64-bit doesn't seem to have the issue; at >> least the .deb builds, anyway. >> >> I get the same wrong-arch warnings as did the original poster: >> ... >> gcc-4.2 -mmacosx-version-min=10.6 -L../../src/.libs/ -arch x86_64 >> -arch i386 -arch ppc -bundle -undefined dynamic_lookup > > That's a lot of off-target -arch for the fink world, which sure > explains the warnings: > >> ld: warning: in /sw32/lib/librrd.dylib, file was built for i386 which >> is not the architecture being linked (x86_64) >> ld: warning: in /sw32/lib/librrd.dylib, file was built for i386 which >> is not the architecture being linked (ppc) > > and why dlopen of the fat .bundle could fail to find symbols in the > single-arch .dylib (not sure how it knows what arch to load, but > obviously not he "right" one for this fink arch config): > >> t/base....Can't load >> '/sw32/src/fink.build/rrd-pm5100-1.3.8-1/rrdtool-1.3.8/bindings/perl-shared/blib/arch/auto/RRDs/RRDs.bundle' >> for module RRDs: >> dlopen(/sw32/src/fink.build/rrd-pm5100-1.3.8-1/rrdtool-1.3.8/bindings/perl-shared/blib/arch/auto/RRDs/RRDs.bundle, >> 2): Symbol not found: _rrd_clear_error >> Referenced from: >> /sw32/src/fink.build/rrd-pm5100-1.3.8-1/rrdtool-1.3.8/bindings/perl-shared/blib/arch/auto/RRDs/RRDs.bundle >> Expected in: flat namespace >> ... >> >> and librrd.dylib seems sane enough: >> >> $ file /sw32/lib/librrd.dylib >> /sw32/lib/librrd.dylib: Mach-O dynamically linked shared library i386 > > So librrd is fine, rrd-pmXXX build process is broken. I bet it's > because CompileScript hard-codes rather than using the default perl > build script fragments...loses the magic that makes single-arch Work.
I committed 1.3.8-3 yesterday, which is likely to build on all architectures now. It's not likely to pass self-tests on i386, but rumor is that there is just something fundamentally broken with perl-tests on this arch, unrelated to this package. So actually, would be great to hear that it actually does work (real-world-test), or if not exactly what's busted. dan -- Daniel Macks dma...@netspace.org http://www.netspace.org/~dmacks ------------------------------------------------------------------------------ ThinkGeek and WIRED's GeekDad team up for the Ultimate GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the lucky parental unit. See the prize list and enter to win: http://p.sf.net/sfu/thinkgeek-promo _______________________________________________ Fink-users mailing list Fink-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/fink-users
