[Wien] lapw1 error "GTFNAM: exactly one command line argument
has to be given"
Peter Blaha
pblaha at theochem.tuwien.ac.at
Thu Jun 30 08:27:51 CEST 2005
Compare the gtfnam.F routines of SRC_lapw0 and SRC_lapw1. Maybe you have an
older WIEN-version somewhere ?
The essential lines look like:
....
iarg=iargc()
!modifications for pgi-mpi compiler
if(iarg.lt.1) goto 900
iarg=1
CALL GETARG(iarg,DEFFN)
....
I see some comment for a pgi compiler!
Eventually you may want to put a print*,iarg before the if statement.
> Thanks, Peter. However,
> Quite strange. Aparently, gtfnam works, otherwise lapw0 should not work
> either.
> Are you running in parallel or sequential ?
> *******************
> >We are running in sequential.
>
> try x lapw1 (-up / -c if you have got a case.in1c file or spinpol.)
> or lapw1(c) (up)lapw1.def
> Does this work ?
> ********************
> >We have tried but the same error message.
>
> Otherwise recompile with -O0 -Mfreeform only (I seem to remember that
> -fast did not work (or produced wrong results ?) with pgi 6.0 compiler
> (on Intel Xeon system), while it was fine with pgi 5.2 ???)
> Anyway, at least for Intel systems the ifort is much faster than pgi
> (don't know about Opterons).
> **********************
> >This makes no difference, either. We use pgi 5.2, I think.
> Do you have any other suggestions?
> Best Regards
>
> 2005/6/29, Peter Blaha <pblaha at theochem.tuwien.ac.at>:
> >
> > Quite strange. Aparently, gtfnam works, otherwise lapw0 should not work
> > either.
> > Are you running in parallel or sequential ?
> >
> > try x lapw1 (-up / -c if you have got a case.in1c file or spinpol.)
> > or lapw1(c) (up)lapw1.def
> >
> > Does this work ?
> >
> > Otherwise recompile with -O0 -Mfreeform only (I seem to remember that
> > -fast did not work (or produced wrong results ?) with pgi 6.0 compiler
> > (on Intel Xeon system), while it was fine with pgi 5.2 ???)
> > Anyway, at least for Intel systems the ifort is much faster than pgi
> > (don't know about Opterons).
> >
> > Regards
> >
> > >
> > > We are running wien2k_05 on a machine of AMD Operton 2.4GHz,64-bit with
> > > Turbo Linux 8.0 for AMD64 system and the pgi (64-bit ) compiler. The
> > > compiler progress is OK, however, when we run the scf cycle,after the
> > lapw0,
> > > an error occured in the lapw1 progrss with the message :" GTFNAM:
> > exactly
> > > one command line argument has to be given ". We dont know what it means
> > and
> > > how to deal with this problem, can anybody give a suggestion?Thanks very
> > > much.
> > >
> > > the comlication conditions and hardware are as follows:
> > > AMD Operton 2.4GHz, 64-bit
> > > Turbo Linux 8.0 for AMD64
> > > pgi (64-bit) compiler
> > > Recommended options for system linuxpgi are:
> > > Compiler options -Mfreeform -fast -Kieee
> > > Linker Flags: -L../SRC_lib
> > > Preprocessor flags: '-DPrarllel'
> > > R_LIB (LAPACK+BLAS): -llapack_lapw -lblas
> > > Current settings:
> > > O Compiler options: -tp=k8-64 -fast -Mfreeform -Mcache_align
> > > L Linker Flags: -L/home/user/sjtusmlt/.pgi/linux86-64/5.2/libso
> > > -L/home/user/sjtusmlt/.acml/pgi64/lib
> > > P Preprocessor flags '-DParallel'
> > > R R_LIB (LAPACK+BLAS): -lacml -lpthread -lm -lg2c -lpgc
> > >
> >
> > P.Blaha
> > --------------------------------------------------------------------------
> > Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
> > Phone: +43-1-58801-15671 FAX: +43-1-58801-15698
> > Email: blaha at theochem.tuwien.ac.at WWW: http://info.tuwien.ac.at/theochem/
> > --------------------------------------------------------------------------
> > _______________________________________________
> > Wien mailing list
> > Wien at zeus.theochem.tuwien.ac.at
> > http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> >
>
P.Blaha
--------------------------------------------------------------------------
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-15671 FAX: +43-1-58801-15698
Email: blaha at theochem.tuwien.ac.at WWW: http://info.tuwien.ac.at/theochem/
--------------------------------------------------------------------------
More information about the Wien
mailing list