[Mpi3-rma] Fix for win_allocate_shared

Torsten Hoefler htor at illinois.edu
Fri Sep 23 13:10:30 CDT 2011

On Fri, Sep 23, 2011 at 12:40:43PM -0500, Pavan Balaji wrote:
> On 09/23/2011 12:39 PM, Torsten Hoefler wrote:
>> It clarifies semantics for separate (the first part) and the advice to
>> implementers makes the semantic restriction explicit.
> What I mean is -- how will an MPI implementation on non-cache-coherent  
> system be able to expose shared memory with this restriction?
Our decision was: Either expose non-cacheable memory or not at all.

Well, the decision is not final, it's very easy to add a rule that says
that the outcome of conflicting remote/local load/store accesses is


PS: Sorry for my sluggish response, we're all stuck on an island with
internet issues ... ;-)

 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-rma mailing list