[CIG-SEISMO] gfortran or ifort?

[Pieyre Le Loher]

Hello,

the code works fine with ifort and gfortran. If you get an error running
"./configure FC=gfortran MPIFC=mpif90" it is maybe because of the gfortran
default flags of SPECFEM3D (for instance if you're version of gfortran is
old). In order to check it, you can change these flags in
SPECFEM3D/flags.guess.

Best regards.
Pieyre

>
> Hello,
>
> The code works fine with any Fortran compiler.
> (or at least it should... if not, please let us know)
>
> Best regards,
> Dimitri.
>
> On 06/29/2010 05:34 PM, 胡元鑫 wrote:
>> Hello,
>> Because the gfortran was bound to MPICH2 of cluster in National High
>> Performence Computing Center (XI'an), China, I must use "./configure
>> FC=gfortran MPIFC=mpif90" instead of "./configure FC=ifort
>> MPIFC=mpif90".  The former passed and created files normally, but the
>> latter got error. The systems administrator told me that they would not
>> change the fortran compiler bound to MPICH2. For that reason, I must use
>> gfortran to compile the SPECFEM3D_SESAME codes. I got some information
>> via Google that the code that works correctly under ifort might behave
>> differently when compiled under a different compiler. Are the computed
>> results different when codes will be compiled under gfortran?
>> Best regards,
>> Hu yuanxin
>
> --
> Dimitri Komatitsch - dimitri.komatitsch aT univ-pau.fr
> Professor, University of Pau, Institut universitaire de France,
> CNRS and INRIA Magique3D, France   http://www.univ-pau.fr/~dkomati1
>