[MPI3 Fortran] MPI non-blocking transfer

[Craig Rasmussen]

On Feb 10, 2009, at 4:48 PM, N.M. Maclaren wrote:

>> You don't really want Fortran to treat buffers like C does as that   
>> will slow down all of your Fortran program.
>
> Please don't be ridiculous.  There is not a simple choice between  
> VOLATILE
> and 'C semantics' - as has been pointed out at length, the precise  
> semantic
> match to MPI non-blocking transfer buffers is the ASYNCHRONOUS  
> attribute.
>

The intent was to remind our C friends that the reason that Fortran  
seems so "odd" in comparison to C is that it can make assumptions that  
improve performance.  Because C can't really know which of its  
variables may be aliased, I believe its true that it has to flush all  
registers at every call site.  Fortran doesn't.

Cheers,
Craig