[mpich-discuss] undefined symbol: MPIR_Thread

Anthony Chan chan at mcs.anl.gov
Mon Feb 7 15:31:01 CST 2011


I just talked to one of our parallel IO experts, it seems your HDF5 library
installed from ubuntu could have hidden MPI dependence.  Could you rebuild
HDF5 with mpich2-1.3.1, and your ccsm.exe afterward ?  If not, you can't use
your own version of mpich2, but the whole stack of software from ubuntu...

A.Chan

----- Original Message -----
> Your unresolved MPIR_Thread error message from loader suggests that
> something
> IN your ccsm.exe depends on libmpich.so, but your ldd output on
> ccsm.exe
> does not show any such dependence. Now you are saying ccsm.exe does
> not even fork/spawn any other executables, I can't think of any other
> ways that could cause unresolved symbol error from loader.
> 
> A.Chan
> 
> ----- Original Message -----
> > The component models are all compiled into and started from a single
> > executable, ccsm.exe,
> > they can either be run sequentially using the same set of mpi tasks
> > or
> > concurrently on disjoint sets of tasks.
> >
> > --
> >
> >
> > -------------------------------------------------------------
> > FWO Post-doctoral Research Fellow
> > Department of Earth& Environmental Sciences
> > Physical and Regional Geography Research Group
> > Regional climate studies
> >
> > Celestijnenlaan 200E
> > 3001 Heverlee (Leuven)
> > BELGIUM
> >
> > Tel: + 32 16 326424
> > Fax: + 32 16 322980
> >
> > http://geo.kuleuven.be/aow/
> > www.kuleuven.be/geography
> > -------------------------------------------------------------
> >
> > _______________________________________________
> > mpich-discuss mailing list
> > mpich-discuss at mcs.anl.gov
> > https://lists.mcs.anl.gov/mailman/listinfo/mpich-discuss


More information about the mpich-discuss mailing list