Re: Clarification of close semantics
Nick Maclaren (nmm1@cus.cam.ac.uk)
Fri, 28 Feb 1997 22:13:02 +0000
> The definition of MPI_Close says "The user is responsible
> for ensuring that all outstanding requests associated with
> fh have completed before calling MPI_CLOSE." (See p217,
> lines 1-2 of the Feb 7 draft.)
>
> This clearly applies to outstanding requests on the node
> that's calling close, but what about outstanding requests
> on the same open file on other nodes? ...
But MPI_CLOSE is a collective routine. The way that I read it, the
processes will call MPI_CLOSE and lock up, until the last has called
in. At some time after that, they will be released. If I am right,
is there still a problem?
Nick Maclaren,
University of Cambridge Computer Laboratory,
New Museums Site, Pembroke Street, Cambridge CB2 3QG, England.
Email: nmm1@cam.ac.uk
Tel.: +44 1223 334761 Fax: +44 1223 334679