Re: comments

Rajeev Thakur (thakur@mcs.anl.gov)
Mon, 13 Jan 1997 10:24:50 -0600

> > Regarding cancelling generalized requests, 137/45 should say, as in
> > MPI-1, that MPI marks the request as "cancelled" not "complete". The
> > user is still required to call MPI_Test/MPI_Wait.
>
> The use of complete was vague. I meant complete in a different
> sense. I changed it to:
>
> MPI marks the request as complete (as if MPI_GR_MARK_COMPLETE was
> called)
>
> What I mean is that the complete_fn is now eligible to execute.

Is the user required to call MPI_Test/MPI_Wait after
canceling a GR, as with regular requests?

> > Also, after a call to
> > MPI_Cancel, the request either completes entirely or no part of it
> > gets completed, but nothing in between. The statement in parenthesis on
> > line 47 is a bit misleading.
>
> I agree that this is desirable but am not sure we want to require
> this. I was hoping the text following covered this. I wanted to say
> that it is a good idea to clean up after oneself but that nothing in
> MPI requires this for a GR. Is this ok?

It is fine with me either way; I was thinking more in terms of
consistency with the behavior of regular requests in MPI-1.

Rajeev