[MPI3 Fortran] [Fwd: Library-based ASYNCHRONOUS I/O and SYNC MEMORY]

[Jeff Squyres]

FWIW, my $0.02 is that it would be great if Fortran allowed temporary  
aliases of buffers.  It could be clearly marked where aliases begin  
and end.  It would also be great if these aliases could be on a per- 
buffer basis -- an all-or-nothing approach like SYNC MEM is good, but  
per-buffer granularity would be better.  I don't need to touch any  
compiler/RTL data -- I just want the compiler/runtime/whatever to let  
me have the user's buffer "in the background" for a while.


On Sep 5, 2008, at 9:15 PM, Dan Nagle wrote:

> Hello,
>
> On Sep 5, 2008, at 2:24 PM, Jeff Squyres wrote:
>>
>> It would be nice to have libraries be able to have asynchronous IO  
>> in Fortran, but I don't pretend to be anything close enough to a  
>> Fortran expert to be able to engage in the debate.  :-\
>>
>
> One issue is that this may involve trying to have
> the external library manipulate data structures
> that are internal to the compiler/RTL.  That might,
> in the view of some vendors, involve releasing
> trade secrets and/or proprietary design details.
>
> In any case, from a standards point of view, it's
> difficult to define, since the internal compiler/RTL
> data structures are not specified by the standard
> (even if, sometimes, there's only One Good Way to do
> something).
>
> -- 
> Cheers!
>
> Dan Nagle
>
>
>
>
> _______________________________________________
> mpi3-fortran mailing list
> mpi3-fortran at lists.mpi-forum.org
> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-fortran


-- 
Jeff Squyres
Cisco Systems