<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; color: rgb(0, 0, 0); font-size: 14px; font-family: Calibri, sans-serif;">
<div>
<div>
<div>Daniel,</div>
<div><br>
</div>
<div>Do you have a test program that shows these errors?</div>
<div><br>
</div>
<div>The algorithms are a first cut, for now, but we are hoping to optimize them in 2016 (together with one of our vendor partners, who will be contributing code for it).</div>
<div><br>
</div>
<div>Thanks,</div>
<div><br>
</div>
<div> -- Pavan</div>
<div>
<div id="MAC_OUTLOOK_SIGNATURE"></div>
</div>
</div>
</div>
<div><br>
</div>
<span id="OLK_SRC_BODY_SECTION">
<div style="font-family:Calibri; font-size:12pt; text-align:left; color:black; BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1pt solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<span style="font-weight:bold">From: </span>Daniel Ibanez <<a href="mailto:dan.a.ibanez@gmail.com">dan.a.ibanez@gmail.com</a>><br>
<span style="font-weight:bold">Reply-To: </span>"<a href="mailto:devel@mpich.org">devel@mpich.org</a>" <<a href="mailto:devel@mpich.org">devel@mpich.org</a>><br>
<span style="font-weight:bold">Date: </span>Saturday, October 24, 2015 at 5:05 PM<br>
<span style="font-weight:bold">To: </span>"<a href="mailto:devel@mpich.org">devel@mpich.org</a>" <<a href="mailto:devel@mpich.org">devel@mpich.org</a>><br>
<span style="font-weight:bold">Subject: </span>[mpich-devel] MPI_Dist_graph_create self-edge<br>
</div>
<div><br>
</div>
<div>
<div>
<div dir="ltr">Hello,
<div><br>
</div>
<div>Using MPICH 3.1.4, I'm getting internal assertions of this kind</div>
<div>of when using MPI_Dist_graph_create :</div>
<div><br>
</div>
<div>Assertion failed in file src/mpi/topo/dist_gr_create.c at line 223: s_rank >= 0<br>
</div>
<div>Assertion failed in file src/mpi/topo/dist_gr_create.c at line 195: sources[i] < comm_size<br>
</div>
<div><br>
</div>
<div>I've checked that ranks passed in are in the proper range,</div>
<div>rather I think the issue is caused by requesting an edge</div>
<div>from rank 0 to itself.</div>
<div>(Another hint is that its non-deterministic which assertion</div>
<div>I get; it depends on the order in which the ranks get scheduled</div>
<div>by the OS, so there is a bit of a bug in the implementation,</div>
<div>or at least no check for self-edges).</div>
<div><br>
</div>
<div>Does the MPI standard allow for self-edges in these graphs ?</div>
<div><br>
</div>
<div>Thank you,</div>
<div><br>
</div>
<div>P.S. - I'll throw in a bigger question while I'm at it: are</div>
<div>MPI_Dist_graph_create and MPI_Neighbor_alltoallv</div>
<div>implemented with optimally scalable algorithms ?</div>
<div>I'm betting my scalability on them being roughly</div>
<div>O(log(P)) where P is communicator size, assuming</div>
<div>neighborhood sizes and message sizes are constant.</div>
</div>
</div>
</div>
</span>
</body>
</html>