[mpiwg-tools] Profiling for MPI_SIZEOF?
m.a.hermanns at grs-sim.de
Thu May 15 15:17:46 CDT 2014
isn't MPI_SIZEOF also named as a function that is allowed to be
implemented as a macro?
Also, the call is likely to be so short (in duration) that you'd have a
huge overhead in intercepting it.
I would also be ok with proposed solution #1.
On 15.05.14 20:42, Jeff Squyres (jsquyres) wrote:
> Question for the Fortran and Tools WGs: How does a tool intercept calls to the Fortran-only MPI_SIZEOF function? (MPI-3.0 p621)
> I ask for the following reasons:
> 1. Unlike all other MPI functions, this function is keyed off the *type* of the 1st argument (not the *value* of the first argument).
> 2. Without support for TYPE(*), DIMENSION(..) (which no compiler supports yet), an MPI implementation has to provide N overloaded versions of MPI_SIZEOF to be able to determine the type of the first argument.
> 3. However, unlike how we specified the Fortran routine names in the MPI-3.0 errata so that they could be intercepted by PMPI, there are no specified specific subroutine names for the overloaded MPI_SIZEOF routines.
> In reality, I'm assuming no tool will ever care about intercepting MPI_SIZEOF (at least, they haven't complained in the several years that Open MPI/MPICH have provided it). But it's an inconsistency in the standard.
> I see two obvious solutions -- perhaps there are more:
> 1. Define that MPI_SIZEOF is not able to be profiled.
> 2. For compilers that do not support TYPE(*), DIMENSION(..), define some kind of pattern of specific procedure names for the overloaded instances of MPI_SIZEOF.
> I'm in favor of #1.
Jülich Aachen Research Alliance,
High Performance Computing (JARA-HPC)
German Research School for Simulation Sciences GmbH
Phone: +49 241 80 99753
Fax: +49 241 80 6 99753
email: m.a.hermanns at grs-sim.de
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 4664 bytes
Desc: S/MIME Cryptographic Signature
More information about the mpiwg-tools