<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
</head>
<body>
<p>Hi Matthew,</p>
<p>we did it with PetscSFCreateInverseSF !</p>
<p>It is working well without overlap, so we can go forward with
this and compute the overlap afterward with
DMPlexDistributeOverlap.</p>
<p>Thanks,</p>
<p>Eric<br>
</p>
<p><br>
</p>
<div class="moz-cite-prefix">On 2021-07-20 10:39 p.m., Eric
Chamberland wrote:<br>
</div>
<blockquote type="cite"
cite="mid:7b11445b-4d50-20d4-7c25-6cb2eab043b6@giref.ulaval.ca">
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
<p><br>
</p>
<div class="moz-cite-prefix">On 2021-07-14 6:42 p.m., Matthew
Knepley wrote:<br>
</div>
<blockquote type="cite"
cite="mid:CAMYG4GkfZyvrc+jNOhLbqqG=qyb6v7ViY2_4mOyaqT_=U2FTpg@mail.gmail.com">
<meta http-equiv="content-type" content="text/html;
charset=UTF-8">
<div dir="ltr">
<div class="gmail_quote"><br>
<div>Ah, there was a confusion of intent. GlobalToNatural()
is for people that want data transformed back into the
original</div>
<div>order. I thought that was what you wanted. If you just
want mesh points in the original order, we give you the</div>
<div>transformation as part of the output of
DMPlexDistribute(). The migrationSF that is output maps
the original point to</div>
<div>the distributed point. You run it backwards to get the
original ordering.</div>
<div><br>
</div>
<div> Thanks,</div>
<div><br>
</div>
<div> Matt </div>
</div>
</div>
</blockquote>
<p>Hi,</p>
<p>that seems to work better! However, if I understand well the
migrationSF is giving information on the originating process
where the elements have been migrated from.</p>
<p>Is there a PETSc way to either:</p>
<p>1) send back the information to the originating process
(somewhat "inverting" the migrationSF) ? So I can retrieve the
"partitioning array" (just like the "part" parameter in
ParMETIS_V3_PartMeshKway) on the sender process.<br>
</p>
<p>or<br>
</p>
<p>2) Have the pre-migrationSF: I mean I would like to extract the
"where are the elements going to be sent?" (again like "part"
parameter)</p>
<p>If not, I can always build the communication myself...<br>
</p>
<p>Thanks,</p>
<p>Eric</p>
<p><br>
</p>
<p><br>
</p>
<p><br>
</p>
</blockquote>
<pre class="moz-signature" cols="72">--
Eric Chamberland, ing., M. Ing
Professionnel de recherche
GIREF/Université Laval
(418) 656-2131 poste 41 22 42</pre>
</body>
</html>