Thanks. Please also apply the attached patch (0004) to fix a trivial bug. The other patch (0003) will make this test XFAIL. You should not apply it yet, but that's where I'm going... Jeff On Thu, Apr 23, 2015 at 10:44 AM, Junchao Zhang <[email protected]> wrote:

I will add your patch. Thanks.

--Junchao Zhang

On Thu, Apr 23, 2015 at 12:31 PM, Jeff Hammond <[email protected]> wrote:

...

Patches for test/mpi/f90/f90types/createf90types.c that test MPI_Type_create_f90_complex for 64b floats are attached.

If these patches are accepted, I want to extend them to support the case of REAL128 (http://fortranwiki.org/fortran/show/Real+precision), which is part of Fortran 2008 and therefore should be supported by MPICH. Tests for REAL128 will fail since MPICH does not support it yet, but I'm hoping to inspire this to change. And I might end up doing it myself if I have to.

Thanks,

Jeff

-- Jeff Hammond [email protected] http://jeffhammond.github.io/

_______________________________________________ To manage subscription options or unsubscribe: https://lists.mpich.org/mailman/listinfo/devel

_______________________________________________ To manage subscription options or unsubscribe: https://lists.mpich.org/mailman/listinfo/devel

-- Jeff Hammond [email protected] http://jeffhammond.github.io/