(j3.2006) [Fwd: Re: [MPI3 Fortran] Feedback from Fortran J3 meeting]

[Aleksandar Donev]

On Wednesday 28 May 2008 10:57, Bill Long wrote:

> If either Craig or Aleks can find any example of a current compiler that
> moves code involving references or definitions of variables with either
> the volatile or asynchronous attributes across ANY call to an external
> routine, then we might need to look further. ?
Wow, what a contradiction. YOU said:

On Tuesday 27 May 2008 13:54, Bill Long wrote:
> Just a reminder that specifying VOLATILE for the actual and
> corresponding dummy arguments is only to prevent copy-in/copy-out of the
> argument to the MPI routine.  Preventing copy-in/copy-out is the goal
> here. The other, normal aspects of VOLATILE are not relevant for this
> example.

Craig and I are arguing that the reason VOLATILE/ASYNCHRONOUS is needed is 
more than just preventing copy in/out---it also inhibits code motion.

So are you actually agreeing with us???

Dan is arguing that VOLATILE is not the right thing and it does the wrong 
thing.

Anyway, please give a coherent statement as to what exactly this argument is 
about?

Best,
Aleks