[mpich-discuss] "undefined references"

Wesley Bland wbland at mcs.anl.gov
Mon Jul 8 12:25:57 CDT 2013


We're not going to be able to help you debug other packages here. You're going to need to talk to the Siesta people about this.

On Jul 8, 2013, at 12:15 PM, Hana Milani <hana.milani at yahoo.com> wrote:

> I have also tried to install serial Siesta and "make" gen basis. Here is the error I get:
> 
> alloc.o basis_types.o precision.o recipes.o parallel.o parsing.o xcmod.o basis_io.o  chemical.o atm_transfer.o atm_types.o atmparams.o old_atmfuncs.o radial.o io.o paste.o  flib_spline.o bessph.o  m_filter.o basis_specs.o atom.o memoryinfo.o memory.o periodic_table.o pseudopotential.o pxf.o dot.o xc.o  atom_options.o arw.o sys.o timer.o  xml.o m_walltime.o gen-basis.o nag.o libfdf.a    dc_lapack.a liblapack.a libblas.a 
> f95: error: dc_lapack.a: No such file or directory
> f95: error: liblapack.a: No such file or directory
> f95: error: libblas.a: No such file or directory
> make: *** [gen-basis] Error 1
> 
> why should it ask for liblapack when I have installed siesta in parallel with absolutely n change in arch.make !!!!
> 
> 
> 
> From: Jeff Hammond <jeff.science at gmail.com>
> To: Hana Milani <hana.milani at yahoo.com>; discuss at mpich.org 
> Sent: Monday, July 8, 2013 8:56 PM
> Subject: Re: [mpich-discuss] "undefined references"
> 
> I looked through the code and suspect Siesta is at fault but since
> they rolled their own MPI F90 wrappers and I know very little about
> F90, I cannot debug it.
> 
> You will see that there are no wrappers for send, recv, bcast or
> reduce (the functions for which you get errors) in ./Src/MPI/mpi.F.
> There are references to these in ./Src/MPI/mpi__*.f90 but I don't
> understand what's going on very well.  It also appears that these
> files are from a prototype implementation of the F90 module for MPI
> 1.1 written in 1996.  I don't know if that was before the MPI standard
> defined F90 bindings, but it surely does now and applications should
> use those instead.
> 
> The other thing you can do is turn off MPI in the utility code since
> it might not be necessary there.
> 
> Jeff
> 
> On Mon, Jul 8, 2013 at 11:09 AM, Hana Milani <hana.milani at yahoo.com> wrote:
> > yes, apologies
> >
> > ________________________________
> > From: Hana Milani <hana.milani at yahoo.com>
> > To: Pavan Balaji <balaji at mcs.anl.gov>
> > Sent: Monday, July 8, 2013 7:53 PM
> >
> > Subject: Re: [mpich-discuss] "undefined references"
> >
> > yes Pavan, But it's a long time I have not received any response
> >
> >
> > ________________________________
> > From: Pavan Balaji <balaji at mcs.anl.gov>
> > To: Hana Milani <hana.milani at yahoo.com>; discuss at mpich.org
> > Sent: Monday, July 8, 2013 7:03 PM
> >
> > Subject: Re: [mpich-discuss] "undefined references"
> >
> >
> > This doesn't look like an MPI error.  Did you contact the authors of
> > whatever package you are trying to compile?
> >
> >  -- Pavan
> >
> > On 07/08/2013 08:25 AM, Hana Milani wrote:
> >> Hello,
> >>
> >> I have compiled a code (siesta) by mpich2, blacs, lapack, blas and
> >> scalapack. It works fine under grid computation. But when I want to
> >> "make" a utility, I receive the following errors:
> >>
> >> [Home at localhost Gen-basis]$ make
> >> /usr/local/bin/mpif90          -o gen-basis \
> >>          alloc.o basis_types.o precision.o recipes.o parallel.o
> >> parsing.o xcmod.o basis_io.o  chemical.o atm_transfer.o atm_types.o
> >> atmparams.o old_atmfuncs.o radial.o io.o paste.o  flib_spline.o
> >> bessph.o  m_filter.o basis_specs.o atom.o memoryinfo.o memory.o
> >> periodic_table.o pseudopotential.o pxf.o dot.o xc.o  atom_options.o
> >> arw.o sys.o timer.o  xml.o m_walltime.o gen-basis.o nag.o libfdf.a
> >> /home/Home/Packages/scalapack-2.0.2/libscalapack.a
> >> /home/Home/Packages/BLACS/LIB/blacsCinit_MPI-LINUX-0.a
> >> /home/Home/Packages/BLACS/LIB/blacsF77init_MPI-LINUX-0.a
> >> /home/Home/Packages/BLACS/LIB/blacs_MPI-LINUX-0.a
> >> /home/Home/Packages/lapack-3.4.2/liblapack.a
> >> /home/Home/Packages/BLAS/blas_LINUX.a
> >> alloc.o: In function `print_report':
> >>
> >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/alloc.F90:1705:
> >> undefined reference to `__mpi__r8_s_MOD_mpi_reduce_t'
> >>
> >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/alloc.F90:1707:
> >> undefined reference to `__mpi__r8_v_MOD_mpi_reduce_t'
> >>
> >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/alloc.F90:1711:
> >> undefined reference to `__mpi__integer_s_MOD_mpi_bcast_t'
> >>
> >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/alloc.F90:1723:
> >> undefined reference to `__mpi__r8_s_MOD_mpi_recv_t'
> >>
> >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/alloc.F90:1718:
> >> undefined reference to `__mpi__r8_s_MOD_mpi_send_t'
> >> alloc.o: In function `__alloc_MOD_alloc_report':
> >>
> >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/alloc.F90:318:
> >> undefined reference to `__mpi__integer_s_MOD_mpi_bcast_t'
> >>
> >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/alloc.F90:319:
> >> undefined reference to `__mpi__integer_s_MOD_mpi_bcast_t'
> >>
> >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/alloc.F90:320:
> >> undefined reference to `__mpi__character_s_MOD_mpi_bcast_t'
> >> xcmod.o: In function `__xcmod_MOD_setxc':
> >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/xcmod.F:136:
> >> undefined
> >> reference to `__mpi__integer_s_MOD_mpi_bcast_t'
> >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/xcmod.F:138:
> >> undefined
> >> reference to `__mpi__character_v_MOD_mpi_bcast_t'
> >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/xcmod.F:140:
> >> undefined
> >> reference to `__mpi__character_v_MOD_mpi_bcast_t'
> >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/xcmod.F:142:
> >> undefined
> >> reference to `__mpi__r8_v_MOD_mpi_bcast_t'
> >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/xcmod.F:144:
> >> undefined
> >> reference to `__mpi__r8_v_MOD_mpi_bcast_t'
> >> memory.o: In function `printmemory_':
> >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/memory.F:144:
> >> undefined reference to `__mpi__integer_s_MOD_mpi_allreduce_t'
> >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/memory.F:167:
> >> undefined reference to `__mpi__integer_sv_MOD_mpi_gather_t'
> >> timer.o: In function `timer1_':
> >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/timer.F:117:
> >> undefined
> >> reference to `__mpi__r8_s_MOD_mpi_allreduce_t'
> >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/timer.F:164:
> >> undefined
> >> reference to `__mpi__r8_v_MOD_mpi_reduce_t'
> >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/timer.F:139:
> >> undefined
> >> reference to `__mpi__r8_v_MOD_mpi_reduce_t'
> >> collect2: error: ld returned 1 exit status
> >> make: *** [gen-basis] Error 1
> >>
> >> I will be really grateful if you kindly help me to solve this problem.
> >>
> >>
> >> _______________________________________________
> >> discuss mailing list    discuss at mpich.org
> >> To manage subscription options or unsubscribe:
> >> https://lists.mpich.org/mailman/listinfo/discuss
> >>
> >
> > --
> > Pavan Balaji
> > http://www.mcs.anl.gov/~balaji
> >
> >
> >
> >
> >
> > _______________________________________________
> > 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
> 
> 
> _______________________________________________
> discuss mailing list     discuss at mpich.org
> To manage subscription options or unsubscribe:
> https://lists.mpich.org/mailman/listinfo/discuss

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mpich.org/pipermail/discuss/attachments/20130708/6d27be7a/attachment.html>


More information about the discuss mailing list