I got the following:
(cd ./built/powerpc-darwin-gfortran && gfortran -dynamiclib
-install_name /sw/lib/gildas/darwin-gfortran/lib/libgsys.so sysc.o
cfc.o gag_trace.o sic_util.o gmessage-c.o gsys-message-c.o
gmessage-c-no-f.o gsys-mod.o gsys-interfaces.o ambigs.o cd.o copy.o
date.o decode.o extrema.o forme.o gagout.o gsys-message.o gtlgtr.o
hashing.o open.o parse.o precess.o putmsg.o string.o sysfor.o tape.o
gpackage.o gmessage.o \
-o
/sw/src/fink.build/gildas-201104f-3/gildas-src-apr11f/integ/powerpc-darwin-gfortran/lib/libgsys.so
-L/sw/src/fink.build/gildas-201104f-3/gildas-src-apr11f/integ/powerpc-darwin-gfortran/lib
-L/System/Library/Frameworks/Accelerate.framework/Frameworks/vecLib.framework/Versions/Current/
-L/sw/lib -L/usr/X11R6/lib -L/sw/lib/python2.6/config -lpython2.6 -lpng
-lz -lfftw3 -lfftw3f -lm)
/usr/bin/ld: common symbols not allowed with MH_DYLIB output format with
the -multi_module option
copy.o definition of common _ourpointerref_ (size 16)
collect2: ld returned 1 exit status
make[3]: ***
[/sw/src/fink.build/gildas-201104f-3/gildas-src-apr11f/integ/powerpc-darwin-gfortran/lib/libgsys.so]
Error 1
rm built/powerpc-darwin-gfortran/forme.f90
built/powerpc-darwin-gfortran/gsys-message.f90
built/powerpc-darwin-gfortran/extrema.f90
built/powerpc-darwin-gfortran/gmessage-c-no-f.o
built/powerpc-darwin-gfortran/gmessage-c.o
built/powerpc-darwin-gfortran/sysc.o
built/powerpc-darwin-gfortran/gsys-mod.f90
built/powerpc-darwin-gfortran/forme.o
built/powerpc-darwin-gfortran/cd.f90
built/powerpc-darwin-gfortran/gsys-message.o
built/powerpc-darwin-gfortran/hashing.f90
built/powerpc-darwin-gfortran/gsys-mod.o
built/powerpc-darwin-gfortran/gsys-interfaces.f90
built/powerpc-darwin-gfortran/cd.o
built/powerpc-darwin-gfortran/date.f90
built/powerpc-darwin-gfortran/parse.f90
built/powerpc-darwin-gfortran/gtlgtr.o
built/powerpc-darwin-gfortran/sic_util.o
built/powerpc-darwin-gfortran/date.o
built/powerpc-darwin-gfortran/extrema.o
built/powerpc-darwin-gfortran/parse.o
built/powerpc-darwin-gfortran/gsys-message-c.o
built/powerpc-darwin-gfortran/hashing.o
built/powerpc-darwin-gfortran/gag_trace.o
built/powerpc-darwin-gfortran/string.f90
built/powerpc-darwin-gfortran/sysfor.f90
built/powerpc-darwin-gfortran/ambigs.f90
built/powerpc-darwin-gfortran/string.o
built/powerpc-darwin-gfortran/gsys-interfaces.o
built/powerpc-darwin-gfortran/copy.f90
built/powerpc-darwin-gfortran/cfc.o
built/powerpc-darwin-gfortran/gagout.f90
built/powerpc-darwin-gfortran/gpackage.f90
built/powerpc-darwin-gfortran/decode.f90
built/powerpc-darwin-gfortran/gmessage.f90
built/powerpc-darwin-gfortran/tape.f90
built/powerpc-darwin-gfortran/sysfor.o
built/powerpc-darwin-gfortran/ambigs.o
built/powerpc-darwin-gfortran/putmsg.f90
built/powerpc-darwin-gfortran/copy.o
built/powerpc-darwin-gfortran/open.f90
built/powerpc-darwin-gfortran/precess.f90
built/powerpc-darwin-gfortran/gagout.o
built/powerpc-darwin-gfortran/gpackage.o
built/powerpc-darwin-gfortran/gmessage.o
built/powerpc-darwin-gfortran/tape.o
built/powerpc-darwin-gfortran/gtlgtr.f90
built/powerpc-darwin-gfortran/putmsg.o
built/powerpc-darwin-gfortran/open.o
built/powerpc-darwin-gfortran/precess.o
built/powerpc-darwin-gfortran/decode.o
make[2]: *** [default-install] Error 1
And I put a full log at
http://akh.users.finkproject.org/finklogs/logfiles/fink-build-log_gildas_201104f-3_2011.07.28-20.44.46
On Thu, Jul 28, 2011 at 9:22 AM, Sébastien Maret
<[email protected]> wrote:
>
> Le 27 juil. 2011 à 16:48, Alexander Hansen a écrit :
>
>> -----BEGIN PGP SIGNED MESSAGE-----
>> Hash: SHA1
>>
>> On 7/24/11 11:53 AM, Sébastien Maret wrote:
>>>
>>> Le 24 juil. 2011 à 17:24, Alexander Hansen a écrit :
>>>
>>>> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
>>>>
>>>> On 7/24/11 10:25 AM, Sébastien Maret wrote:
>>>>>
>>>>> Le 23 juil. 2011 à 14:17, Alexander Hansen a écrit :
>>>>>
>>>>>>> 10.4 is no longer officially supported, but maybe the
>>>>>>> maintainer will fix it--or somebody else can.
>>>>>>>
>>>>>>> I get what may be the same error from gildas-201104f-2 with
>>>>>>> the following setup:
>>>>>>>
>>>>>>> Package manager version: 0.30.2 Distribution version:
>>>>>>> selfupdate-cvs Fri Jul 22 18:46:43 2011, 10.4, powerpc Trees:
>>>>>>> stable/main stable/crypto unstable/main unstable/crypto
>>>>>>> local/main local/injected Xcode Version: 2.5-1
>>>>>>>
>>>>>>> ... gfortran -Wl,-commons,use_dylibs
>>>>>>> built/powerpc-darwin-gfortran/hlp2tex.o -o
>>>>>>> /sw/src/fink.build/gildas-201104f-2/gildas-src-apr11f/integ/powerpc-darwin-gfor
>>>>>>>
>>>>>>>
>>>>>>
>>>>>>>
>>>>>>>
>>>>
>>>>>>>
>> tran/bin/hlp2tex \
>>>>>>>
>>>>>>> -L/sw/src/fink.build/gildas-201104f-2/gildas-src-apr11f/integ/powe
>>>>>>>
>>>>>>>
>>>>
>>>>>>>
>> rpc-darwin-gfortran/lib
>>>>>>> -L/System/Library/Frameworks/Accelerate.framework/Framew
>>>>>>> orks/vecLib.framework/Versions/Current/ -L/sw/lib
>>>>>>> -L/usr/X11R6/lib -L/sw/lib/pyt hon2.6/config -lpython2.6
>>>>>>> -lutil -lpng -lz -lfftw3 -lfftw3f -L/sw/lib -lgtk-x1 1-2.0
>>>>>>> -lgdk-x11-2.0 -latk-1.0 -lgio-2.0 -lpangoft2-1.0
>>>>>>> -lgdk_pixbuf-2.0 -lpangoc airo-1.0 -lcairo -lpango-1.0
>>>>>>> -lfreetype -lfontconfig -lgobject-2.0 -lgmodule-2.0
>>>>>>> -lgthread-2.0 -lglib-2.0 -lintl -lm /usr/bin/ld: unknown
>>>>>>> flag: -commons ...
>>>>>>>
>>>>>>> 10.4's "ld" doesn't understand -commons except on 64-bit
>>>>>>> builds.
>>>>>
>>>>> I don't think that Gildas can work without this flag. Is there an
>>>>> equivalent in 10.4's ld?
>>>>>
>>>>> Sébastien
>>>>
>>>> According to 10.4's ld manpage:
>>>>
>>>> - -commons treatment (64-bit only) Specifies how common symbols
>>>> (tentative defintions) from object files interact with dynamic
>>>> libraries. treatment can be: ignore_dylibs, use_dylibs, or
>>>> error. The default is ignore_dylibs which means the static
>>>> linker will use a common defintion from an object file even if
>>>> a true definition exisits in a dynamic library. If you want your
>>>> code to use a dynamic library definition, then add the extern
>>>> keyword to your tenta- tive definition (e.g. change int foo; to
>>>> extern int foo; ). The treatment use_dylibs means a definition
>>>> form a dynamic library should override a common symbol in an
>>>> object file.
>>>>
>>>> Note, the 32-bit linker always uses this treatment.
>>>>
>>>> The treatment error means the linker should abort whenever if finds
>>>> a common symbol in an object file and an external definition with
>>>> the same name in a dynamic library.
>>>>
>>>> So it sounds to me like if you leave that out, the 32-bit linker
>>>> will default to the same behavior as -commons,use_dylib .
>>>
>>> OK. I can I fix this ? Should I create a gildas-10.4-32bit.info
>>> package ?
>>>
>>> Sébastien
>>
>> I have played around with both the unstable and stable tree versions of
>> gildas on 10.4, but I'm not sure how to proceed. I've posted build logs at
>>
>> http://akh.users.finkproject.org/finklogs/logfiles/
>>
>> and the .info files are in my experimental:
>>
>> http://fink.cvs.sf.net/viewvc/fink/experimental/alexkhansen/10.4/finkinfo/3rdparty/
>
> Thanks, I think I've found the problem. Could you please try to build gildas
> on 10.4 ppc with these .info and .patch files:
>
> https://public.me.com/sebastien.maret/
>
> Sébastien
>
>
>
>
>
>
>
>
------------------------------------------------------------------------------
Got Input? Slashdot Needs You.
Take our quick survey online. Come on, we don't ask for help often.
Plus, you'll get a chance to win $100 to spend on ThinkGeek.
http://p.sf.net/sfu/slashdot-survey
_______________________________________________
Fink-beginners mailing list
[email protected]
List archive:
http://news.gmane.org/gmane.os.apple.fink.beginners
Subscription management:
https://lists.sourceforge.net/lists/listinfo/fink-beginners
