Re: GRs and request completion

Dick Treumann (treumann@kgn.ibm.com)
Tue, 06 May 1997 17:17:31 -0400

Steve Huss-Lederman wrote:

>
> The call informs \MPI/ that the generalized request \mpiarg{request}
> completes (see definitions in Section~\ref{terms}). A call to
> \mpifunc{MPI\_WAIT(request)} will return and a call to
> \mpifunc{MPI\_TEST(request, flag)} will return \mpiarg{flag=true} only
> after the generalized request completes by a call to
> \mpifunc{MPI\_GREQUEST\_COMPLETE}.

How about:
The call informs \MPI/ that the operations which the generalized
represents are complete. (see definitions in Section~\ref{terms}). A
call to \mpifunc{MPI\_WAIT(request)} will return and a call to
\mpifunc{MPI\_TEST(request, flag)} will return \mpiarg{flag=true} only
after a call to \mpifunc{MPI\_GREQUEST\_COMPLETE} has declared that
these operations are complete.

The rest of Steves changes look fine to me, not because of "completes"
but because they break the juxtaposition of "request" and "complete"
which was present in the earlier wordings. My suggested wording for the
final item is also intended to avoid the use of "complete" in connection
with "request". I think a distinction between "completes" and
"completed" is too subtle for anyone to notice.

Dick

-- 
Dick Treumann                               IBM RS/6000 Division
(Internet) treumann@pok.ibm.com             Poughkeepsie, NY
(VNET)     TREUMANN at KGNVMC               Tel: (914) 433-7846
(internal) treumann@windsurf.pok.ibm.com    Fax: (914) 433-8363