|
[Rivet] installation problems (partially solved?)Holger Schulz holger.schulz at physik.hu-berlin.deThu Jul 24 16:45:06 BST 2008
Andy Buckley wrote: > Holger Schulz wrote: > >> James Monk wrote: >> >>> Hi Holger, >>> >>> I think your AGILe library may actually be built 32 bit, too >>> (otherwise how could the 32 bit generator libs be loaded?). >>> >> I just saw, that the LCG-tag is created by the bootstrap-script using >> 'uname -m' which returns 'i686'. So I guess >> that having a 64 bit processor installed doesn't neccessarily mean that >> your system architecture is 64 bit , too. >> > > Correct - often 32 bit OSes are run on 64 bit hardware. Of course, it's > also possible that the bootstrap script is wrong... it is a bit of a hack! > > >> I am starting from scratch with the latest version of rivet-bootstrap now. >> > > Cool: let us know how it goes. > Allright, now it compiles against cern's libraries. But runnning e.g. 'rivetgun -P lep1.params -g Pythia6:413' results in a segfault. Here is the gdb backtrace: ********************** PYINIT: initialization completed ********************** Program received signal SIGSEGV, Segmentation fault. [Switching to Thread -1208088896 (LWP 10542)] 0x001a1494 in HepMC::IO_HEPEVT::build_production_vertex (this=0x8790d18, i=1, hepevt_particle=@0xbffa84c0, evt=0xbffa8630) at ../HepMC/HEPEVT_Wrapper.h:263 263 double* mydouble = (double*)&hepevt.data[b]; Current language: auto; currently c++ (gdb) bt #0 0x001a1494 in HepMC::IO_HEPEVT::build_production_vertex (this=0x8790d18, i=1, hepevt_particle=@0xbffa84c0, evt=0xbffa8630) at ../HepMC/HEPEVT_Wrapper.h:263 #1 0x001a39ab in HepMC::IO_HEPEVT::fill_next_event (this=0x8790d18, evt=0xbffa8630) at IO_HEPEVT.cc:84 #2 0x005aff30 in AGILe::FPythia::fillEvent () from /afs/ifh.de/user/h/hschulz/Software/lib/libAGILeFPythia.so #3 0x005aff87 in AGILe::FPythia::makeEvent () from /afs/ifh.de/user/h/hschulz/Software/lib/libAGILeFPythia.so #4 0x08053f19 in Rivet::generate () #5 0x08050183 in main () I am using CERN's HepMC of version 2.04.00 Holger > Andy > >
More information about the Rivet mailing list |