(j3.2006) (SC22WG5.3904) [ukfortran] [MPI3 Fortran] MPI non-blocking transfers

Van Snyder Van.Snyder
Thu Jan 22 17:01:24 EST 2009


On Thu, 2009-01-22 at 11:21 -0800, Bill Long wrote:

> This suffers from the same problem as Van's extra argument proposal.
> The buffer may not be visible in the scoping unit of the call to
> MPI_wait.  In addition, it may not be clear which buffer is relevant
> to a particular call.  Nice idea, but I don't think it is practical.

If the extra argument is a pointer that is TKR compatible with the
buffer, the compiler has no choice but not to fiddle with anything
that's TKR compatible and has the TARGET or POINTER attribute, or has
such a subobject.  If the buffer isn't a pointer and isn't a target and
isn't accessible in the scoping unit of the MPI_WAIT call, the compiler
can't possibly do anything nefarious to it.

What other conjectures do you have that might cause the extra argument
approach not to work?






More information about the J3 mailing list