Hmmm...well, a little digging says that we probably didn't do this as thoroughly as we should have :-/

Basically, the "MPI I/O Support" line is tied solely to the disable-mpi-io flag. In the 1.6 series, you only had one option for MPI I/O support, and that was romio. So if you disable romio, then you'll get a "failed" reply to those APIs.

What we should have done is automatically "disable-mpi-io" if no components were built - i.e., if you disable them all from building. In 1.7, that would mean disabling both romio and ompi-io. In 1.6, that would just mean disabling romio.

I'll pass it along to Jeff to see if it can be done - we can certainly change the flag, but maybe not for the 1.6 series

On Feb 13, 2014, at 12:01 PM, Blosch, Edwin L <> wrote:

Why does ompi_info –c say “MPI I/O Support: yes” even though I configured using –disable-io-romio?  If ompi_info is going to tell me MPI I/O is supported, then shouldn’t I expect my test program (attached) to work correctly?  (it doesn’t).
I didn’t disable “built-in” mpi-io, only io-romio.
  --disable-mpi-io        Disable built-in support for MPI-2 I/O, likely
                          because an externally-provided MPI I/O package will
                          be used. Default is to use the internal component
                          system and its specially modified version of ROMIO
  --disable-io-romio      Disable the ROMIO MPI-IO component
configure options used:
+ ./configure --prefix=/applocal/tools/mpi/intel/openmpi-1.6.4 --without-tm --without-sge --without-lsf --without-psm --without-portals --without-elan --without-slurm --without-loadleveler --with-mxm=/opt/mellanox/mxm --with-mxm-lib=/opt/mellanox/mxm/lib --enable-mpirun-prefix-by-default --enable-contrib-no-build=vt --disable-per-user-config-files --disable-io-romio --enable-static CXX=/applocal/intel/composer_xe_2013/bin/icpc CC=/applocal/intel/composer_xe_2013/bin/icc 'CFLAGS=  -O2' 'CXXFLAGS=  -O2' F77=/applocal/intel/composer_xe_2013/bin/ifort 'FFLAGS=-D_GNU_SOURCE -traceback  -O2' FC=/applocal/intel/composer_xe_2013/bin/ifort 'FCFLAGS=-D_GNU_SOURCE -traceback  -O2' 'LDFLAGS= -static-intel'
ompi_info –c output:
           Configured by: bloscel
           Configured on: Tue Jun 11 16:20:00 CDT 2013
          Configure host: mgmt1
                Built by: bloscel
                Built on: Tue Jun 11 16:35:12 CDT 2013
              Built host: mgmt1
              C bindings: yes
            C++ bindings: yes
      Fortran77 bindings: yes (all)
      Fortran90 bindings: yes
Fortran90 bindings size: small
              C compiler: /applocal/intel/composer_xe_2013/bin/icc
     C compiler absolute:
  C compiler family name: INTEL
      C compiler version: 1310.20130514
             C char size: 1
             C bool size: 1
            C short size: 2
              C int size: 4
             C long size: 8
            C float size: 4
           C double size: 8
          C pointer size: 8
            C char align: 1
            C bool align: 1
             C int align: 4
           C float align: 4
          C double align: 8
            C++ compiler: /applocal/intel/composer_xe_2013/bin/icpc
   C++ compiler absolute: none
      Fortran77 compiler: /applocal/intel/composer_xe_2013/bin/ifort
  Fortran77 compiler abs:
      Fortran90 compiler: /applocal/intel/composer_xe_2013/bin/ifort
  Fortran90 compiler abs:
       Fort integer size: 4
       Fort logical size: 4
Fort logical value true: -1
      Fort have integer1: yes
      Fort have integer2: yes
      Fort have integer4: yes
      Fort have integer8: yes
     Fort have integer16: no
         Fort have real4: yes
         Fort have real8: yes
        Fort have real16: no
      Fort have complex8: yes
     Fort have complex16: yes
     Fort have complex32: no
      Fort integer1 size: 1
      Fort integer2 size: 2
      Fort integer4 size: 4
      Fort integer8 size: 8
     Fort integer16 size: -1
          Fort real size: 4
         Fort real4 size: 4
         Fort real8 size: 8
        Fort real16 size: 16
      Fort dbl prec size: 8
          Fort cplx size: 8
      Fort dbl cplx size: 16
         Fort cplx8 size: 8
        Fort cplx16 size: 16
        Fort cplx32 size: 32
      Fort integer align: 1
     Fort integer1 align: 1
     Fort integer2 align: 1
     Fort integer4 align: 1
     Fort integer8 align: 1
    Fort integer16 align: -1
         Fort real align: 1
        Fort real4 align: 1
        Fort real8 align: 1
       Fort real16 align: 1
     Fort dbl prec align: 1
         Fort cplx align: 1
     Fort dbl cplx align: 1
        Fort cplx8 align: 1
       Fort cplx16 align: 1
       Fort cplx32 align: 1
             C profiling: yes
           C++ profiling: yes
     Fortran77 profiling: yes
     Fortran90 profiling: yes
          C++ exceptions: no
          Thread support: posix (MPI_THREAD_MULTIPLE: no, progress: no)
           Sparse Groups: no
            Build CFLAGS: -DNDEBUG -O2 -finline-functions -fno-strict-aliasing -restrict -pthread
          Build CXXFLAGS: -DNDEBUG -O2 -finline-functions -pthread
            Build FFLAGS: -D_GNU_SOURCE -traceback  -O2
           Build FCFLAGS: -D_GNU_SOURCE -traceback  -O2
           Build LDFLAGS: -export-dynamic  -static-intel
              Build LIBS: -lrt -lnsl  -lutil
    Wrapper extra CFLAGS: -pthread
  Wrapper extra CXXFLAGS: -pthread
    Wrapper extra FFLAGS:
   Wrapper extra FCFLAGS:
   Wrapper extra LDFLAGS:
      Wrapper extra LIBS: -ldl  -lm -lnuma  -Wl,--export-dynamic -lrt -lnsl -lutil
  Internal debug support: no
  MPI interface warnings: no
     MPI parameter check: runtime
Memory profiling support: no
Memory debugging support: no
         libltdl support: yes
   Heterogeneous support: no
mpirun default --prefix: yes
         MPI I/O support: yes
       MPI_WTIME support: gettimeofday
     Symbol vis. support: yes
   Host topology support: yes
          MPI extensions: affinity example
   FT Checkpoint support: no (checkpoint thread: no)
     VampirTrace support: no
        MPI_MAX_INFO_KEY: 36
        MPI_MAX_INFO_VAL: 256
       MPI_MAX_PORT_NAME: 1024
users mailing list