Ah -- make dist when using icc/ifort was the magic combination that fails (it 
wouldn't fail when using gcc/gfortran).

Ok, I see the problem -- fix coming shortly.  Thanks for finding it!


On Feb 3, 2014, at 7:44 AM, Mike Dubman <mi...@dev.mellanox.co.il> wrote:

> more info:
> 
> to reproduce, "./autogen.sh && ./configure && make dist" on rhel 6.4
> 
> 
> On Mon, Feb 3, 2014 at 12:39 PM, Mike Dubman <mi...@dev.mellanox.co.il> wrote:
> (cd scripts && make  top_distdir=../../../../../openmpi-1.7.4rc4git 
> distdir=../../../../../openmpi-1.7.4rc4git/ompi/mpi/fortran/use-mpi-tkr/scripts
>  \
>      am__remove_distdir=: am__skip_length_check=: am__skip_mode_fix=: distdir)
> make[4]: Entering directory 
> `/hpc/local/benchmarks/hpc-stack-icc/src/install/ompi-v1.7/ompi/mpi/fortran/use-mpi-tkr/scripts'
> make[4]: Leaving directory 
> `/hpc/local/benchmarks/hpc-stack-icc/src/install/ompi-v1.7/ompi/mpi/fortran/use-mpi-tkr/scripts'
> make[3]: Leaving directory 
> `/hpc/local/benchmarks/hpc-stack-icc/src/install/ompi-v1.7/ompi/mpi/fortran/use-mpi-tkr'
>  (cd mpi/fortran/use-mpi-ignore-tkr && make  
> top_distdir=../../../../openmpi-1.7.4rc4git 
> distdir=../../../../openmpi-1.7.4rc4git/ompi/mpi/fortran/use-mpi-ignore-tkr \
>      am__remove_distdir=: am__skip_length_check=: am__skip_mode_fix=: distdir)
> make[3]: Entering directory 
> `/hpc/local/benchmarks/hpc-stack-icc/src/install/ompi-v1.7/ompi/mpi/fortran/use-mpi-ignore-tkr'
> make[3]: Leaving directory 
> `/hpc/local/benchmarks/hpc-stack-icc/src/install/ompi-v1.7/ompi/mpi/fortran/use-mpi-ignore-tkr'
>  (cd mpi/fortran/use-mpi-f08 && make  
> top_distdir=../../../../openmpi-1.7.4rc4git 
> distdir=../../../../openmpi-1.7.4rc4git/ompi/mpi/fortran/use-mpi-f08 \
>      am__remove_distdir=: am__skip_length_check=: am__skip_mode_fix=: distdir)
> make[3]: Entering directory 
> `/hpc/local/benchmarks/hpc-stack-icc/src/install/ompi-v1.7/ompi/mpi/fortran/use-mpi-f08'
> make[3]: Leaving directory 
> `/hpc/local/benchmarks/hpc-stack-icc/src/install/ompi-v1.7/ompi/mpi/fortran/use-mpi-f08'
>  (cd mpi/fortran/use-mpi-f08-desc && make  
> top_distdir=../../../../openmpi-1.7.4rc4git 
> distdir=../../../../openmpi-1.7.4rc4git/ompi/mpi/fortran/use-mpi-f08-desc \
>      am__remove_distdir=: am__skip_length_check=: am__skip_mode_fix=: distdir)
> make[3]: Entering directory 
> `/hpc/local/benchmarks/hpc-stack-icc/src/install/ompi-v1.7/ompi/mpi/fortran/use-mpi-f08-desc'
>   FC       mpi-f08-types.lo
> mpi-f08-types.f90(83): error #6418: This name has already been assigned a 
> data type.   [MPI_SUBARRAYS_SUPPORTED]
>   logical, parameter :: MPI_SUBARRAYS_SUPPORTED = .true.
> ------------------------^
> compilation aborted for mpi-f08-types.f90 (code 1)
> make[3]: *** [mpi-f08-types.lo] Error 1
> make[3]: Leaving directory 
> `/hpc/local/benchmarks/hpc-stack-icc/src/install/ompi-v1.7/ompi/mpi/fortran/use-mpi-f08-desc'
> make[2]: *** [distdir] Error 1
> make[2]: Leaving directory 
> `/hpc/local/benchmarks/hpc-stack-icc/src/install/ompi-v1.7/ompi'
> make[1]: *** [distdir] Error 1
> make[1]: Leaving directory 
> `/hpc/local/benchmarks/hpc-stack-icc/src/install/ompi-v1.7'
> make: *** [dist] Error 2
> 
> 
> _______________________________________________
> devel mailing list
> de...@open-mpi.org
> http://www.open-mpi.org/mailman/listinfo.cgi/devel


-- 
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to: 
http://www.cisco.com/web/about/doing_business/legal/cri/

Reply via email to