Re: caching on MPI handles

Eric Salo (salo@mrjones.engr.sgi.com)
Fri, 18 Oct 1996 15:03:14 -0700

Messages sorted by: [ date ][ thread ][ subject ][ author ]
Next message: Rajeev Thakur: "Re: caching on MPI handles"
Previous message: Rajeev Thakur: "caching on file handles"

> Yes, I mainly care about the delete function. I don't intend to use
> the copy function. So, if we say "when a new datatype is created
> from an old one, the attribute does not get propagated to the new
> one", that would be fine with me.

With communicators, the copy_fn is only invoked for MPI_COMM_DUP, right? Since
we don't have a MPI_TYPE_DUP function, we could at least claim that we're being
consistent. So the next question to ask is, should there be a MPI_TYPE_DUP? I
seem to recall someone asking for this a while back.

-- 
Eric Salo         Silicon Graphics Inc.             "Do you know what the
(415)933-2998     2011 N. Shoreline Blvd, 8U-802     last Xon said, just
salo@sgi.com      Mountain View, CA   94043-1389     before he died?"

Next message: Rajeev Thakur: "Re: caching on MPI handles"
Previous message: Rajeev Thakur: "caching on file handles"