[DFTB-Plus-User] compiling 22.2 - error in compiling MBD section
Carlo Nervi
carlo.nervi at unito.it
Thu Feb 23 14:53:28 CET 2023
Dear Bàlint,
thank you for the prompt reply.
I'm working under Gentoo linux with the following software:
gcc version 9.3.0 (Gentoo 9.3.0-r2 p4), OpenBLAS (zenp) 0.3.12, fftw3
3.5.8, scalapack 2.2
mpirun (Open MPI) 2.0.4
I suspect that this latter is the problem: OpenMPI is not updated to the
latest 4.1.4. Indeed I tested the compilation on another Gentoo with
OpenMPI 4.1.4 (and OpenBlas 0.3.19, gcc Gentoo 12.2.1_p20230121-r1 p10) and
it worked fine (BTW, gcc 12 it seems to me to compile rather slowly).
The fact that dftb+ 22.1 compiled fine while 22.2 not, surprised me.
I'm a little bit refrained from upgrading the whole Gentoo (a lot of work
to do...and compatibility issues may arises with other programs...)
Thank you,
Carlo
Il giorno gio 23 feb 2023 alle ore 09:06 Bálint Aradi <aradi at uni-bremen.de>
ha scritto:
> Dear Carlo,
>
> can you give some more details, especially which compiler and which
> MP-framework (and which versions of those) you use? It seems a to be an
> an mpif08 problem within MBD.
>
> Best regards,
>
> Bálint
>
>
> On 22.02.23 19:27, Carlo Nervi wrote:
> > Dear developers of dftb+ 22.2,
> > I'm compiling dftb+ version 22.2 (github version) under linux gentoo.
> > I switched on (in config.cmake) WITH_MPI, WITH_TBLITE, WITH_MBD and
> > WITH_SDFTD3
> >
> > The compilation with cmake proceed up to the point concerning MBD.
> > It must be pointed out that with 22.1 I had no problems in compiling.
> > Now the following error appears:
> >
> > [ 6%] Built target test_gather
> > [ 6%] Built target test_gatherv
> > [ 8%] Built target scalapackfx
> > [ 8%] Building Fortran object
> > external/mbd/origin/src/CMakeFiles/mbd.dir/mbd_geom.F90.o
> >
> /home/nervi/src/dftb/22.2_dev/dftbplus/external/mbd/origin/src/mbd_geom.F90:77:32:
> >
> > 77 | type(MPI_Comm) :: mpi_comm = MPI_COMM_WORLD
> > | 1
> > Error: Parameter ‘mpi_comm_world’ at (1) has not been declared or is a
> > variable, which does not reduce to a constant expression
> >
> /home/nervi/src/dftb/22.2_dev/dftbplus/external/mbd/origin/src/mbd_geom.F90:153:36:
> >
> > 153 | call MPI_COMM_SIZE(this%mpi_comm, this%mpi_size, ierr)
> > | 1
> > Error: ‘mpi_comm’ at (1) is not a member of the ‘geom_t’ structure; did
> > you mean ‘mpi_size’?
> >
> /home/nervi/src/dftb/22.2_dev/dftbplus/external/mbd/origin/src/mbd_geom.F90:154:36:
> >
> > 154 | call MPI_COMM_RANK(this%mpi_comm, this%mpi_rank, ierr)
> > | 1
> > Error: ‘mpi_comm’ at (1) is not a member of the ‘geom_t’ structure; did
> > you mean ‘mpi_size’?
> >
> /home/nervi/src/dftb/22.2_dev/dftbplus/external/mbd/origin/src/mbd_geom.F90:172:47:
> >
> > 172 | call this%blacs_grid%init(this%mpi_comm%mpi_val)
> > | 1
> > Error: ‘mpi_comm’ at (1) is not a member of the ‘geom_t’ structure; did
> > you mean ‘mpi_size’?
> >
> /home/nervi/src/dftb/22.2_dev/dftbplus/external/mbd/origin/src/mbd_geom.F90:253:90:
> >
> > 253 | call MPI_ALLGATHER(this%exc%code, 1, MPI_INTEGER, codes, 1,
> > MPI_INTEGER, this%mpi_comm, err)
> > |
> > 1
> > Error: ‘mpi_comm’ at (1) is not a member of the ‘geom_t’ structure; did
> > you mean ‘mpi_size’?
> >
> /home/nervi/src/dftb/22.2_dev/dftbplus/external/mbd/origin/src/mbd_geom.F90:256:77:
> >
> > 256 | call MPI_BCAST(this%exc%code, 1, MPI_INTEGER, rank,
> > this%mpi_comm, err)
> > |
> > 1
> > Error: ‘mpi_comm’ at (1) is not a member of the ‘geom_t’ structure; did
> > you mean ‘mpi_size’?
> >
> /home/nervi/src/dftb/22.2_dev/dftbplus/external/mbd/origin/src/mbd_geom.F90:258:83:
> >
> > 258 | this%exc%msg, len(this%exc%msg), MPI_CHARACTER,
> > rank, this%mpi_comm, err &
> > |
> > 1
> > Error: ‘mpi_comm’ at (1) is not a member of the ‘geom_t’ structure; did
> > you mean ‘mpi_size’?
> >
> /home/nervi/src/dftb/22.2_dev/dftbplus/external/mbd/origin/src/mbd_geom.F90:261:89:
> >
> > 261 | this%exc%origin, len(this%exc%origin),
> > MPI_CHARACTER, rank, this%mpi_comm, err &
> > |
> > 1
> > Error: ‘mpi_comm’ at (1) is not a member of the ‘geom_t’ structure; did
> > you mean ‘mpi_size’?
> > gmake[2]: ***
> > [external/mbd/origin/src/CMakeFiles/mbd.dir/build.make:186:
> > external/mbd/origin/src/CMakeFiles/mbd.dir/mbd_geom.F90.o] Error 1
> > gmake[1]: *** [CMakeFiles/Makefile2:4374:
> > external/mbd/origin/src/CMakeFiles/mbd.dir/all] Error 2
> > gmake: *** [Makefile:160: all] Error 2
> >
> > Any suggestion is greatly appreciated.
> > Thank you,
> > Carlo
> >
> > --
> >
> > ------------------------------------------------------------
> > Prof. Carlo Nervicarlo.nervi at unito.it <mailto:carlo.nervi at unito.it>
> Tel:+39 0116707507/8
> > Fax: +39 0116707855 - Dipartimento di Chimica, via
> > P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/ <
> http://lem.ch.unito.it/>
> >
> >
> > _______________________________________________
> > DFTB-Plus-User mailing list
> > DFTB-Plus-User at mailman.zfn.uni-bremen.de
> >
> https://mailman.zfn.uni-bremen.de/cgi-bin/mailman/listinfo/dftb-plus-user
>
> --
> Dr. Bálint Aradi
> Bremen Center for Computational Materials Science, University of Bremen
> http://www.bccms.uni-bremen.de/cms/people/b-aradi/
>
>
> _______________________________________________
> DFTB-Plus-User mailing list
> DFTB-Plus-User at mailman.zfn.uni-bremen.de
> https://mailman.zfn.uni-bremen.de/cgi-bin/mailman/listinfo/dftb-plus-user
>
--
------------------------------------------------------------
Prof. Carlo Nervi carlo.nervi at unito.it Tel:+39 0116707507/8
Fax: +39 0116707855 - Dipartimento di Chimica, via
P. Giuria 7, 10125 Torino, Italy. http://lem.ch.unito.it/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.zfn.uni-bremen.de/pipermail/dftb-plus-user/attachments/20230223/15e2b508/attachment.htm>
More information about the DFTB-Plus-User
mailing list