[petsc-dev] Bad use of defined(MPI_XXX)

[Jed Brown]

Lisandro Dalcin via petsc-dev <petsc-dev at mcs.anl.gov> writes:

> On Sat, 25 May 2019 at 06:20, Smith, Barry F. via petsc-dev <
> petsc-dev at mcs.anl.gov> wrote:
>
>>
>>   MS-MPI has in their include file
>>
>> #define MPI_VERSION     2
>> #define MPI_SUBVERSION  0
>>
>> Their last release was Oct 2018
>> https://docs.microsoft.com/en-us/message-passing-interface/microsoft-mpi-release-notes
>>
>>
> Maybe they are just being honest? Other implementations claim MPI 2.0
> support, but they do not support e.g. MPI_Comm_spawn() and related routines.
>
> I think some important bits like MPI_Reduce_local() have been supported
> since long time ago. Note that v4.2 is from 2014 (
> https://www.microsoft.com/en-us/download/details.aspx?id=41634).
>
> Here you have a summary of what MS MPI adds on top of MPI 2.0 release after
> release starting from v4.2
>
> https://bitbucket.org/mpi4py/mpi4py/src/master/src/lib-mpi/config/msmpi.h

To their credit, MS-MPI is on GitHub now (though judging from activity,
it's either abandoned or development is taking place internally).  In
any case, you can watch this line to see if they declare support beyond
MPI-2.0.

https://github.com/microsoft/Microsoft-MPI/blob/master/src/include/mpi.h#L4060