[MPI3 Fortran] [Fwd: Was ASYNCHRONOUS..., New: ACCESSES ( ) suffix

[Aleksandar Donev]

On Wednesday 10 September 2008 16:00, you wrote:

> The idea behind the accesses(
> ) suffix is to make the associated subroutine be treated a similar
> special way.  
What is wrong with my proposed SYNC attribute? Why do you need to specify 
which specific variables may be "tainted" by which specific waiting 
operation? This adds a hell of a lot of syntax with at best dubiously-defined 
semantics for a reason I am still unable to understand.

> It also gets around the problem of needing to modify the 
> source to declare buffer asynchronous.
My proposed SYNC attribute did the same.

Aleks