RE: [mpi-21] ABI (was: Call for MPI 2.2 and 3.0...)

["Rajeev Thakur" <thakur@xxxxxxxxxxx>]

More than holes in the standard, it is because people (sometimes
unknowingly) assume things not guaranteed by the standard, e.g., code that
relies on MPI_Send being  nonblocking or MPI_Get being blocking. So it's
more often user error.

Rajeev

> -----Original Message-----
> From: owner-mpi-21@xxxxxxxxxxxxx 
> [mailto:owner-mpi-21@xxxxxxxxxxxxx] On Behalf Of Greg Lindahl
> Sent: Wednesday, December 05, 2007 1:19 PM
> To: mpi-21@xxxxxxxxxxxxx
> Subject: Re: [mpi-21] ABI (was: Call for MPI 2.2 and 3.0...)
> 
> On Wed, Dec 05, 2007 at 10:37:18AM -0000, Supalov, Alexander wrote:
> 
> > In other words,
> > an application may not work upon transition to a new MPI. 
> In fact, very
> > often it does not.
> 
> Is there any way to collect more details about such instances, so that
> we can think about possible holes in the standard?
> 
> -- greg
> 
>