[Mpi3-hybridpm] [Mpi3-rma] Alternative Proposal for Shared Memory Support

Torsten Hoefler htor at illinois.edu
Tue Mar 15 22:04:14 CDT 2011

On Tue, Mar 15, 2011 at 09:48:40PM -0500, Rajeev Thakur wrote:
> Also, at the last meeting we had trouble defining the MPI_SHM_Fence
> function. Dave said we won't be able to define it portably for all
> processor architectures wrt memory orderings etc. Someone else pointed
> out that, in Fortran, code can be moved around it. Those issues won't
> go away by effectively renaming MPI_SHM_Fence to MPI_Win_flush. (These
> issues are wrt direct loads/stores to shared memory.)
I know, those issues are the same for a fast implementation on a shared
memory system (one that doesn't do a membar in every put/get/...). So I
think this issue is orthogonal and independent of the interface.


 bash$ :(){ :|:&};: --------------------- http://www.unixer.de/ -----
Torsten Hoefler         | Performance Modeling and Simulation Lead
Blue Waters Directorate | University of Illinois (UIUC)
1205 W Clark Street     | Urbana, IL, 61801
NCSA Building           | +01 (217) 244-7736

More information about the mpiwg-hybridpm mailing list