[Mpi3-rma] [EXTERNAL] Re: request-based ops

Barrett, Brian W bwbarre at sandia.gov
Tue Jun 18 16:31:21 CDT 2013


On 6/17/13 9:38 AM, "Jim Dinan" <james.dinan at gmail.com> wrote:

> Hi Brian, 
> 
> Re: iunlock, ifence --
> 
> It might make sense to look at defining MPI_Win_test for these synchronization
> modes.  Some of the semantics might already be defined for us.

I don't think so.  MPI_WIN_TEST is only defined for completing an access
epoch of generalized active target.  It doesn't actually have the semantics
you want, as it's almost a moving flush.  That is, a valid implementation
would is to ask "does my started counter and ack counter match?", over and
over, moving the target.  There's no splitting of an operation (the
operation is "am i done?").  But MPI_WIN_FLUSH has much different semantics;
you're splitting an operation into a start/complete, which means there's
operation ordering / completion issues that don't exist with MPI_WIN_TEST.

Brian

--
  Brian W. Barrett
  Scalable System Software Group
  Sandia National Laboratories


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mpi-forum.org/pipermail/mpiwg-rma/attachments/20130618/6dd53b0d/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 454 bytes
Desc: not available
URL: <http://lists.mpi-forum.org/pipermail/mpiwg-rma/attachments/20130618/6dd53b0d/attachment-0001.bin>


More information about the mpiwg-rma mailing list