Open MPI logo

Open MPI Development Mailing List Archives

  |   Home   |   Support   |   FAQ   |   all Development mailing list

Subject: Re: [OMPI devel] MPI Forum question?
From: Ralph Castain (rhc_at_[hidden])
Date: 2010-04-30 12:12:55


Guess this has been too upsetting a question - I'll work off-list with the other developers to determine an appropriate OMPI behavior.

On Apr 30, 2010, at 9:28 AM, N.M. Maclaren wrote:

> On Apr 30 2010, Ralph Castain wrote:
>> On Apr 30, 2010, at 6:15 AM, Jeff Squyres wrote:
>>
>>>> MPI quite rightly does not specify this, because the matter is very system- dependent, and it is not possible to return the exit code (or display it) in all environments. Sorry, but that is reality.
>>> Correct -- MPI intentionally does not say what happens after MPI_FINALIZE. MPI intentionally doesn't even specify much about how to start an MPI job (just like Fortran, actually).
>>
>> Frankly, I disagree - I think the standard can and should say something explicit about this situation. It doesn't have to say how we implement it, but it should clearly explain to users what they should expect to see at the end of an MPI job.
>>
>> Guess the real issue is: is the standard written for the general community, or solely for MPI implementers? If the latter, then saying nothing is fine. If the former, then it needs to clearly say something about this question.
>
> No, that's NOT the real issue! The real issue is whether it CAN say
> anything more without harming its portability.
>
> You would have to start by defining what you mean as an "MPI job", and
> I can assure you that it would be very hard to do. You may have used
> only a few, very similar, MPIs and parallel environments, but I have
> used a fairly wide range - and I know that I have only very limited
> experience relative to what is out there!
>
> There are environments where the MPI invokation command completes as
> soon as the task has been submitted to a scheduler. In that case, it
> obviously can't return anything based on the job execution. Should
> such implementations be forbidden?
>
> Regards,
> Nick Maclaren.
>
>
> _______________________________________________
> devel mailing list
> devel_at_[hidden]
> http://www.open-mpi.org/mailman/listinfo.cgi/devel