[mpich2-dev] MPIX vs. MPI etc.

Pavan Balaji balaji at mcs.anl.gov
Sun Aug 19 23:05:47 CDT 2012


On 08/19/2012 11:02 PM, Jeff Hammond wrote:
>> Of course, there are no guarantees with MPIX_ functions and we can do
>> whatever we want with them.  But it's trivial to just remove those
>> functions; in fact, lesser work than just not implementing them.
>
> Look, I would not do it the stupid way, I just want a hard verdict on
> whether or not the stupid way is not permitted or just discouraged.
> You and Bill seem to disagree on this.

It's allowed, but not encouraged.

Everything is allowed with MPIX_.  It's valid (though bone-headed) to 
provide an MPIX_Send function that receives data every 5th time it's 
called, and aborts when a prime number of bytes are used.  There's no 
standard for MPIX_ functions.

-- 
Pavan Balaji
http://www.mcs.anl.gov/~balaji


More information about the mpich2-dev mailing list