[Mpi-forum] MPI-2.2 errata tickets - may be urgent

Fab Tillier ftillier at microsoft.com
Fri Feb 17 13:22:45 CST 2012


Dave Goodell wrote on Fri, 17 Feb 2012 at 10:59:30

<snip...>

> #199 is a very small (sort of) problem w.r.t. our inconsistent IN/OUT/INOUT
> usage.  I'm inclined to say that we should leave this up to the chapter
> committee as well.  Fab could do us all a favor by withdrawing it again.

Already withdrawn, because the IN/OUT/INOUT annotations aren't strictly for that function's parameter use (see MPI_Irecv, for example - the buffer pointer is not really OUT for that function).

Also, #197 is already fixed in the 3.0 standard, so I closed that one too.

> #198 is something that needs to be handled by chapter
> authors/committees, although the correct action to take would need to be
> decided by a beauty contest at the forum level.  My personal view is
> that this ticket is a waste of time.

You're probably right, and tickets 125/126 end up touching a lot of the pointers anyway...  The placement of the * is most certainly a ticket-0 change, if not less than 0 - it's a whitespace change.  I might tackle it when I'm looking for mindless distractions, I just need to know which way to go.  From looking at the standard document, it seems the only inconsistency relates to void* parameters.  Should be easy enough to fix, and probably makes sense to change to "void *" to match other pointer parameters. 

-Fab





More information about the mpi-forum mailing list