<html>
<head>
<style>
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 10pt;
font-family:Verdana
}
</style>
</head>
<body class='hmmessage'>
<br>Rajeev<div><br></div><div>From PNNL it was Sriram Krishnamoorthy.</div><div><br></div><div>Others, if you have up-coming system with unique cache coherence or memory access constrains that you think I should explicitly address, please write me a note.</div><div><br></div><div>Thanks,</div><div>Vinod.</div><div><br></div><div><br>> From: thakur@mcs.anl.gov<br>> To: mpi3-rma@lists.mpi-forum.org<br>> Date: Tue, 9 Dec 2008 11:24:13 -0600<br>> Subject: [Mpi3-rma] Notes from RMA working group concall, Dec 9, 2008<br>> <br>> Attendees:<br>> Rajeev, Darius, Dave Goodell, Argonne<br>> Bill Gropp, UIUC<br>> Rich, Vinod, ORNL<br>> Jesper, Hubert, NEC<br>> Robert Blackmore, Su Huang, IBM<br>> Howard Pritchard, Cray<br>> Christian Bell, Myricom<br>> Someone from PNNL (missed the name)<br>> I may have missed others<br>> <br>> * Discussion on issues with non-cache-coherent systems. Vinod is<br>> preparing a presentation for the next Forum mtg. He needs<br>> input from vendors who have (or will have) systems that are not<br>> cache coherent or have special memory features.<br>> <br>> * Vinod is updating the existing proposal and will send out a new<br>> version before the Forum mtg.<br>> <br>> * Vinod will check with Dan Bonachea if he can attend next mtg and his<br>> thoughts on current proposal.<br>> <br>> * Discussion on how does the target know of completion. Vinod will<br>> clarify further in new version of proposal and FAQ.<br>> <br>> * Long discussion on how to represent the target address. Currently it<br>> is directly provided as a (void *), which won't work<br>> everywhere. There was some consensus that it needs to be an object<br>> of some kind. Care must be taken about scalability issues (don't<br>> want a process to a hold a list of everyone's addresses).<br>> <br>> * Ticket 35 (datatype and Op registration) and 37 (examples describing<br>> synchronization semantics) were briefly discussed. Jesper will make<br>> a presentation on 35 at the Forum mtg. Rajeev said having examples<br>> like in #37 is a good thing; we need to make sure they are correct <br>> (he thinks they are).<br>> <br>> If I have missed anything, please add.<br>> <br>> Rajeev<br>> <br>> _______________________________________________<br>> mpi3-rma mailing list<br>> mpi3-rma@lists.mpi-forum.org<br>> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-rma<br></div></body>
</html>