[mpiwg-rma] Changes to the RMA chapter
robl at mcs.anl.gov
Thu Mar 5 10:26:21 CST 2015
On 03/05/2015 09:19 AM, Balaji, Pavan wrote:
> Hi Rob,
>> On Mar 5, 2015, at 9:01 AM, Rob Latham <robl at mcs.anl.gov> wrote:
>> On 03/04/2015 08:35 PM, Jim Dinan wrote:
>>> I agree, no_locks seems to have a clear local semantic. accumulate_ops
>>> ad accumulate_ordering need some thought from the WG; I think these were
>>> intended to be provided at all processes.
>> if you are going to call it an MPI_Info, then please follow the rules of MPI_Info objects already well established in the I/O chapter. all processes must specify the same value for a given MPI_Info key.
> I didn't realize we say that in the MPI standard. Can you point me to the location where it says that?
Thanks for forcing me to re-scan the standard. MPI_Info appears in a
lot more places than I remembered.
I have conflated "The way ROMIO does it" with "the way the standard
says", and now see the restriction on identical keys must be a ROMIO
thing: page 499 of MPI-3.0:
MPI_FILE_SET_INFO sets new values for the hints of the file
associated with fh. MPI_FILE_SET_INFO is a collective routine. The info
object may be different on each process, but any info entries that an
implementation requires to be the same on all processes must appear with
the same value in each process’s info object.
so you can do whatever you want with MPI_Info objects, apparently.
Mathematics and Computer Science Division
Argonne National Lab, IL USA
More information about the mpiwg-rma