[mpich-discuss] ld: warning: ignoring file /Users/jhammond/eclipse/MPICH2/mpich.git/build-clang/lib/.libs/libmpich.a, file was built for archive which is not the architecture being linked (x86_64): /Users/jhammond/eclipse/MPICH2/mpich.git/build-clang/lib/.libs/libmpich.a

William Gropp wgropp at illinois.edu
Mon Aug 19 20:00:07 CDT 2013


Weird.  I am unable to reproduce the problem even when I specify clang as the compiler on my Macbook.  What if you force 64bit on C and C++?  Or can you tell which specific object files (rather than libraries) have the wrong architecture?

Bill

William Gropp
Director, Parallel Computing Institute
Deputy Director for Research
Institute for Advanced Computing Applications and Technologies
Thomas M. Siebel Chair in Computer Science
University of Illinois Urbana-Champaign




On Aug 18, 2013, at 5:54 PM, Jeff Hammond wrote:

> Hi Bill,
> 
> Complete details can be found here:
> 
> $ ls -alrt >& ls.out ; ../configure CC=clang CXX=clang++ FC=false F77=false --disable-fc --disable-f77 --enable-threads --with-device=ch3:nemesis --prefix=/Users/jhammond/eclipse/MPICH2/git/install-clang >& config.out ; make V=1 >& make.log 
> 
> https://dl.dropboxusercontent.com/u/58983082/ls.out (trivial - just to show I start from scratch)
> https://dl.dropboxusercontent.com/u/58983082/config.log
> https://dl.dropboxusercontent.com/u/58983082/config.out
> https://dl.dropboxusercontent.com/u/58983082/make.log
> 
> Thanks,
> 
> Jeff
> 
> 
> On Sun, Aug 18, 2013 at 11:28 AM, Jeff Hammond <jeff.science at gmail.com> wrote:
> Did config.log not show up? I see the attachment in my outbox. 
> 
> This is the output of make the second time. Everything is fine until this step but I can attach the whole thing if necessary. 
> 
> I tried a dozen or so attempts, all from a completely empty directory (including deleting .libs and such). 
> 
> Also, I ran autogen.sh after my clean pull yesterday. 
> 
> Best, 
> 
> Jeff
> 
> Sent from my iPhone
> 
> On Aug 18, 2013, at 9:37 AM, William Gropp <wgropp at illinois.edu> wrote:
> 
>> Jeff,
>> 
>> Can you send the configure output as well?  It is strange that no files were built.  Does this happen if you do a make clean first?
>> 
>> Bill
>> 
>> William Gropp
>> Director, Parallel Computing Institute
>> Deputy Director for Research
>> Institute for Advanced Computing Applications and Technologies
>> Thomas M. Siebel Chair in Computer Science
>> University of Illinois Urbana-Champaign
>> 
>> 
>> 
>> 
>> On Aug 17, 2013, at 10:26 PM, Jeff Hammond wrote:
>> 
>>> I'm very happy to accept that this is user error, either due to direct stupidity or indirectly by virtue of something I've done with macports (or using macports in the first place), but I can't figure out how to solve it.  I've made a number of queries to the google oracle to no avail.
>>> 
>>> I tried the stackoverflow suggestion of -m64 as well as -arch x86_64, but neither worked.
>>> 
>>> Any suggestions?
>>> 
>>> Thanks,
>>> 
>>> Jeff
>>> 
>>> Jeffs-MacBook-Pro:build-clang jhammond$ make
>>> make  all-recursive
>>> Making all in /Users/jhammond/eclipse/MPICH2/mpich.git/build-clang/src/mpl
>>> make[2]: Nothing to be done for `all'.
>>> Making all in /Users/jhammond/eclipse/MPICH2/mpich.git/build-clang/src/openpa
>>> Making all in src
>>> make  all-am
>>> Making all in test
>>> make[3]: Nothing to be done for `all'.
>>> make[3]: Nothing to be done for `all-am'.
>>> Making all in src/mpi/romio
>>> Making all in src/pm/hydra
>>> Making all in ../../mpl
>>> make[3]: Nothing to be done for `all'.
>>> Making all in tools/topo/hwloc/hwloc
>>> Making all in src
>>> make[4]: Nothing to be done for `all'.
>>> Making all in include
>>> make[4]: Nothing to be done for `all'.
>>> make[4]: Nothing to be done for `all-am'.
>>> Making all in .
>>> Making all in .
>>>   CCLD     src/env/mpichversion
>>> ld: warning: ld: warning: ignoring file /Users/jhammond/eclipse/MPICH2/mpich.git/build-clang/src/mpl/.libs/libmpl.a, file was built for archive which is not the architecture being linked (x86_64): /Users/jhammond/eclipse/MPICH2/mpich.git/build-clang/src/mpl/.libs/libmpl.aignoring file lib/.libs/libpmpich.a, file was built for archive which is not the architecture being linked (x86_64): lib/.libs/libpmpich.a
>>> 
>>> ld: warning: ignoring file lib/.libs/libmpich.a, file was built for archive which is not the architecture being linked (x86_64): lib/.libs/libmpich.a
>>> ld: warning: ignoring file /Users/jhammond/eclipse/MPICH2/mpich.git/build-clang/lib/.libs/libmpich.a, file was built for archive which is not the architecture being linked (x86_64): /Users/jhammond/eclipse/MPICH2/mpich.git/build-clang/lib/.libs/libmpich.a
>>> Undefined symbols for architecture x86_64:
>>>   "_MPIR_Version_CC", referenced from:
>>>       _main in mpichversion.o
>>>   "_MPIR_Version_CXX", referenced from:
>>>       _main in mpichversion.o
>>>   "_MPIR_Version_F77", referenced from:
>>>       _main in mpichversion.o
>>>   "_MPIR_Version_FC", referenced from:
>>>       _main in mpichversion.o
>>>   "_MPIR_Version_configure", referenced from:
>>>       _main in mpichversion.o
>>>   "_MPIR_Version_date", referenced from:
>>>       _main in mpichversion.o
>>>   "_MPIR_Version_device", referenced from:
>>>       _main in mpichversion.o
>>>   "_MPIR_Version_string", referenced from:
>>>       _main in mpichversion.o
>>> ld: symbol(s) not found for architecture x86_64
>>> clang: error: linker command failed with exit code 1 (use -v to see invocation)
>>> make[2]: *** [src/env/mpichversion] Error 1
>>> make[1]: *** [all-recursive] Error 1
>>> make: *** [all] Error 2
>>> 
>>> -- 
>>> Jeff Hammond
>>> jeff.science at gmail.com
>>> <config.log>_______________________________________________
>>> discuss mailing list     discuss at mpich.org
>>> To manage subscription options or unsubscribe:
>>> https://lists.mpich.org/mailman/listinfo/discuss
>> 
> 
> 
> 
> -- 
> Jeff Hammond
> jeff.science at gmail.com

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mpich.org/pipermail/discuss/attachments/20130819/6bcaa3b8/attachment.html>


More information about the discuss mailing list