<div dir="ltr">Little did you know, you'd still be holding these meetings 4 years later. :)<div><br></div><div>Meeting notes:</div><div><br></div><div>In attendance: Dan and Jim</div><div><br></div><div>Open items for endpoints:</div>
<div><br></div><div>** Interface for freeing endpoints communicators:</div><div><br></div><div>Current, we use MPI_Comm_free to free endpoints communicators, and we extend the MPI_Comm_free semantics to allows one thread to call this routine for each endpoint. If we didn't have this extension to Comm_free, you can create endpoints, e.g. in MPI_THREAD_SERIALIZED mode, but you can't free them. The Forum dislikes this new semantic and as suggested that we remove it.</div>
<div><br></div><div>Dan suggested a third approach where we leave the Comm_free semantic unchanged and add a new function, MPI_Comm_free_endpoints(MPI_Comm comm_handles[], int my_num_ep), that has symmetry with Comm_create_endpoints. An endpoints communicator can be freed with either Comm_free or Comm_free_endpoints. They differ in their concurrency requirement.</div>
<div><br></div><div>** MPI_ALIASED result from comm_compare:</div><div><br></div><div>We need to define aliased, on page 237 for comm_compare. We may also need to define it for group_compare. Here's suggested text for comm_compare (changes highlighted):</div>
<div>
<div class="" title="Page 267">
<div class="">
<div class="">
<p><span style="font-size:10pt;font-family:CMSS10">MPI</span><span style="font-size:10pt;font-family:CMTT10">_</span><span style="font-size:10pt;font-family:CMSS10">IDENT </span><span style="font-size:11pt;font-family:CMR10">results if and only if </span><span style="font-size:11pt;font-family:CMSS10">comm1 </span><span style="font-size:11pt;font-family:CMR10">and </span><span style="font-size:11pt;font-family:CMSS10">comm2 </span><span style="font-size:11pt;font-family:CMR10">are handles for the same object (<span style="background-color:rgb(255,255,0)">same rank, </span>identical
groups<span style="background-color:rgb(255,255,0)">,</span> and same contexts). </span><span style="background-color:rgb(255,255,0)"><span style="font-size:10pt;font-family:CMSS10">MPI</span><span style="font-size:10pt;font-family:CMTT10">_</span><span style="font-size:10pt;font-family:CMSS10">ALIASED </span><span style="font-size:11pt;font-family:CMR10">results if </span><span style="font-size:11pt;font-family:CMSS10">comm1 </span><span style="font-size:11pt;font-family:CMR10">and </span><span style="font-size:11pt;font-family:CMSS10">comm2 </span></span><span style="font-size:11pt;font-family:CMR10"><span style="background-color:rgb(255,255,0)">are handles for different ranks in the same communicator (different ranks, identical groups, and same contexts).</span> </span><span style="font-size:10pt;font-family:CMSS10">MPI</span><span style="font-size:10pt;font-family:CMTT10">_</span><span style="font-size:10pt;font-family:CMSS10">CONGRUENT </span><span style="font-size:11pt;font-family:CMR10">results if the underlying groups are identical
in constituents and rank order; these communicators diff</span><span style="font-size:11pt;font-family:CMR10">er only by context. </span><span style="font-size:10pt;font-family:CMSS10">MPI</span><span style="font-size:10pt;font-family:CMTT10">_</span><span style="font-size:10pt;font-family:CMSS10">SIMILAR
</span><span style="font-size:11pt;font-family:CMR10">results if the group members of both communicators are the same but the rank order diff</span><span style="font-size:11pt;font-family:CMR10">ers.
</span><span style="font-size:10pt;font-family:CMSS10">MPI</span><span style="font-size:10pt;font-family:CMTT10">_</span><span style="font-size:10pt;font-family:CMSS10">UNEQUAL </span><span style="font-size:11pt;font-family:CMR10">results otherwise. </span></p>
</div></div></div></div><div> ~Jim.</div></div><div class="gmail_extra"><br><br><div class="gmail_quote">On Mon, Apr 7, 2014 at 12:24 PM, Balaji, Pavan <span dir="ltr"><<a href="mailto:balaji@anl.gov" target="_blank">balaji@anl.gov</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Folks,<br>
<br>
We were supposed to have a hybrid telecon today. Unfortunately, when I created my calendar invite 4 years ago, I decided that March 2014 was a great end date for the telecons and didn’t have it show up on my calendar.<br>
<br>
Jim and Dan met. Jim will email out notes.<br>
<br>
Sorry about that.<br>
<br>
— Pavan<br>
<br>
_______________________________________________<br>
mpiwg-hybridpm mailing list<br>
<a href="mailto:mpiwg-hybridpm@lists.mpi-forum.org">mpiwg-hybridpm@lists.mpi-forum.org</a><br>
<a href="http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-hybridpm" target="_blank">http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-hybridpm</a><br>
</blockquote></div><br></div>