[Rivet] Rivet segfaults (Mac OS X)

[David Grellscheid]

Hi Stefan,

One thing to exclude would be a mixed compiler environment used to link 
libraries (e.g. a recent XCode update / newly installed gcc from 
elsewhere?) where Rivet was originally compiled with one version, but 
now another one is found instead or a similar mismatch between Rivet and 
its dependencies.

Can you try the DYLD_PRINT_RPATHS, DYLD_PRINT_LIBRARIES{,_POST_LAUNCH} 
environment variables (man dyld) and check the paths carefully if 
they're all as you expect.

Thanks,

   David


On 07/03/2016 14:03, Stefan Richter wrote:
> Dear Rivet authors,
> can you help me with the below problem? Rivet segfaults when I’m trying to run analyses.
>
> Thank you very much,
> Stefan
>
>
>
> If I don’t attach any analyses to the run, it works:
>
> ---------------------------
>> rivet  ./original_run/original.hepmc
> Rivet 2.4.0 running on machine pb-d-128-141-166-175.cern.ch (x86_64) at 2016-03-07 13:54:48
> Reading events from './original_run/original.hepmc'
> Finished event loop at 2016-03-07 13:54:48
> Cross-section = 6.087660e+00 pb
>
> Rivet.Analysis.Handler: INFO  Finalising analyses
> Rivet.Analysis.Handler: INFO  Processed 50 events
>
> The MCnet usage guidelines apply to Rivet: see http://www.montecarlonet.org/GUIDELINES
> Please acknowledge plots made with Rivet analyses, and cite arXiv:1003.0694 (http://arxiv.org/abs/1003.0694)
>
> Rivet run completed at 2016-03-07 13:54:48, time elapsed = 0:00:00
> ---------------------------
>
>
>
> With an analysis, it segfaults:
>
> ---------------------------
>> rivet ./original_run/original.hepmc -a MC_ZZJETS
> Rivet 2.4.0 running on machine pb-d-128-141-166-175.cern.ch (x86_64) at 2016-03-07 13:57:57
> Segmentation fault: 11
> ---------------------------
>
>
>
> The DYLD_LIBRARY_PATH and PYTHONPATH seem to be fine, they include the relevant locations.
>
>
>
> GDB output:
>
> ---------------------------
> Rivet 2.4.0 running on machine pb-d-128-141-166-74.cern.ch (x86_64) at 2016-03-03 20:14:54
> Rivet.Analysis.Handler: WARN  Analysis 'MC_ZZJETS_CUSTOMIZED' is unvalidated: be careful, it may be broken!
>
> Program received signal SIGSEGV, Segmentation fault.
> 0x00007fff8cdd4774 in __cxxabiv1::__si_class_type_info::has_unambiguous_public_base(__cxxabiv1::__dynamic_cast_info*, void*, int) const () from /usr/lib/libc++abi.dylib
> (gdb) backtrace
> #0  0x00007fff8cdd4774 in __cxxabiv1::__si_class_type_info::has_unambiguous_public_base(__cxxabiv1::__dynamic_cast_info*, void*, int) const () from /usr/lib/libc++abi.dylib
> #1  0x000000010221e6f9 in __dynamic_cast () from /sw/lib/gcc5/lib/libstdc++.6.dylib
> #2  0x0000000104619457 in Rivet::MC_ZZJETS_CUSTOMIZED::init() () from /Users/stefanrichter/Code/workspace/Rivet/analyses/RivetMC_ZZJETS_CUSTOMIZED.so
> #3  0x00000001031c556a in Rivet::AnalysisHandler::init(HepMC::GenEvent const&) () from /Users/stefanrichter/Software/local/lib/libRivet.dylib
> #4  0x00000001031850c9 in Rivet::Run::init(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, double) ()
>     from /Users/stefanrichter/Software/local/lib/libRivet.dylib
> #5  0x0000000102202060 in __pyx_pw_5rivet_4core_3Run_9init(_object*, _object*, _object*) () from /Users/stefanrichter/Software/local/lib/python2.7/site-packages/rivet/core.so
> #6  0x00000001000b26af in PyEval_EvalFrameEx () from /sw/lib/python2.7/config/libpython2.7.dylib
> #7  0x00000001000aac22 in PyEval_EvalCodeEx () from /sw/lib/python2.7/config/libpython2.7.dylib
> #8  0x00000001000aa616 in PyEval_EvalCode () from /sw/lib/python2.7/config/libpython2.7.dylib
> #9  0x00000001000d96c4 in PyRun_FileExFlags () from /sw/lib/python2.7/config/libpython2.7.dylib
> #10 0x00000001000d9241 in PyRun_SimpleFileExFlags () from /sw/lib/python2.7/config/libpython2.7.dylib
> #11 0x00000001000ecfb8 in Py_Main () from /sw/lib/python2.7/config/libpython2.7.dylib
> #12 0x00007fff8e5605fd in start () from /usr/lib/system/libdyld.dylib
> #13 0x00007fff8e5605fd in start () from /usr/lib/system/libdyld.dylib
> #14 0x0000000000000000 in ?? ()
> ---------------------------
>
> Looks like a problem in dynamic linking? Should I run 'otool -L' on the Rivet libraries?
>
> _______________________________________________
> Rivet mailing list
> Rivet at projects.hepforge.org
> https://www.hepforge.org/lists/listinfo/rivet
>