[Mpi3-subsetting] MPI subsetting: charting the way forward atatelecon next week?

[Richard Treumann]

Hi Martin

Assertions are not MPI  implementation specific.  There are features of the
MPI standard and semantic guarantees of the MPI standard that a few
applications depend on and many do not. Some of these standard MPI features
are costly in memory footprint or damage performance.

A program author will be able to use a predefined assertion to declare that
this application does not require a specific feature. The MPI
implementation is then free to use optimizations that are not allowable if
the feature or guarantee must be maintained just in case it is needed.

It may not be simple to explain the meaning of each assertion but the
meaning is 100% rooted in the MPI standard and is not different from one
implementation to the next.

Dick Treumann  -  MPI Team/TCEM
IBM Systems & Technology Group
Dept 0lva / MS P963 -- 2455 South Road -- Poughkeepsie, NY 12601
Tele (845) 433-7846         Fax (845) 433-8363

<snip>

> .............      Also, if these constants
> are really implementation specific, does it make sense to have
> them in the MPI standard? Each vender will want their own set
> (and rightfully so) and the burden is then on the programmer to
> know all of the different options and understand the subtle
> differences (and we have to document them all in the standard).
<snip>
>
> Martin
>
<snip>





* 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mpi-forum.org/pipermail/mpi3-subsetting/attachments/20080620/6785e27a/attachment.html>