<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=us-ascii">
</head>
<body style="word-wrap: break-word; -webkit-nbsp-mode: space; line-break: after-white-space;" class="">
Bump - is this question too complex or too stupid? :-)
<div class=""><br class="">
</div>
<div class="">I'm now looking again on tests which are still failing in knepley/fix-plex-interpolate-sf-new.</div>
<div class=""><br class="">
</div>
<div class="">There are several ones failing with</div>
<div class=""> Parallel partitioning of uninterpolated meshes not supported</div>
<div class="">So I'll just remove them for now.</div>
<div class=""><br class="">
</div>
<div class="">One test (m_impls_plex_tests-ex18_2b) is failing just because of diff. I'll just check if the current output is correct and update the output file.</div>
<div class=""><br class="">
</div>
<div class="">But these three</div>
<div class=""> dm_impls_plex_tests-ex18_6_tet_nsize-4interpolate-parallel_faces-2,2,2</div>
<div class=""> dm_impls_plex_tests-ex18_6_tet_nsize-4interpolate-parallel_faces-1,3,5</div>
<div class=""> dm_impls_plex_tests-ex18_6_tet_nsize-4interpolate-parallel_faces-3,4,7</div>
<div class="">fail with</div>
<div class=""> point SF contains X but not Y from its cone</div>
<div class="">Can it also be a problem of partitioning, and not interpolation?</div>
<div class="">So should we just for now disable ctetgen without interpolation and remove those tests?</div>
<div class=""><br class="">
</div>
<div class="">If it is the case, knepley/fix-plex-interpolate-sf-new should be merged soon, just after removing the failing tests.</div>
<div class=""><br class="">
</div>
<div class="">Thanks</div>
<div class=""><br class="">
</div>
<div class="">Vaclav</div>
<div class=""><br class="">
<div><br class="">
<blockquote type="cite" class="">
<div class="">On 31 Jan 2019, at 15:33, Hapla Vaclav <<a href="mailto:vaclav.hapla@erdw.ethz.ch" class="">vaclav.hapla@erdw.ethz.ch</a>> wrote:</div>
<br class="Apple-interchange-newline">
<div class="">
<div style="word-wrap: break-word; -webkit-nbsp-mode: space; line-break: after-white-space;" class="">
Matt, your commit
<div class=""><a href="https://bitbucket.org/petsc/petsc/commits/ffbd61634109cd55a4e1889a4f100c5a1bdbfef1" class="">https://bitbucket.org/petsc/petsc/commits/ffbd61634109cd55a4e1889a4f100c5a1bdbfef1</a></div>
<div class="">says</div>
<div class="">Parallel partitioning of uninterpolated meshes not supported.</div>
<div class=""><br class="">
</div>
<div class="">So should I now take it as a not-yet-implemented feature?</div>
<div class=""><br class="">
</div>
<div class="">Does it really concern only this order of operations?</div>
<blockquote style="margin: 0 0 0 40px; border: none; padding: 0px;" class="">
<div class="">1) load sequentially</div>
<div class="">2) distribute</div>
<div class="">3) redistribute in parallel</div>
<div class="">4) interpolate in parallel</div>
</blockquote>
<div class=""><br class="">
</div>
<div class="">So the following three use cases</div>
<div class="">I) </div>
<blockquote style="margin: 0 0 0 40px; border: none; padding: 0px;" class="">
<div class="">
<div class="">1) load sequentially</div>
</div>
<div class="">
<div class="">2) interpolate sequentially</div>
</div>
<div class="">
<div class="">
<div class="">3) distribute</div>
</div>
</div>
<div class="">
<div class="">4) redistribute in parallel</div>
</div>
</blockquote>
<div class="">II)</div>
<blockquote style="margin: 0 0 0 40px; border: none; padding: 0px;" class="">
<div class="">
<div class="">1) load sequntially</div>
</div>
<div class="">
<div class="">2) distribute</div>
</div>
<div class="">
<div class="">3) interpolate in parallel</div>
</div>
<div class="">
<div class="">4) redistribute in parallel</div>
</div>
</blockquote>
<div class="">
<div class="">III)</div>
</div>
<blockquote style="margin: 0 0 0 40px; border: none; padding: 0px;" class="">
<div class="">
<div class="">1) load in parallel</div>
</div>
<div class="">2) interpolate in parallel</div>
<div class="">
<div class="">3) redistribute in parallel</div>
</div>
</blockquote>
<div class="">can still be considered supported in general?</div>
<div class=""><br class="">
</div>
<div class="">I mean not regarding the issues with the SF which still seem to affect parallel interpolation (<a href="https://bitbucket.org/petsc/petsc/branch/knepley/fix-plex-interpolate-sf" class="">https://bitbucket.org/petsc/petsc/branch/knepley/fix-plex-interpolate-sf</a>).</div>
<div class="">BTW I now rebased this branch onto master as <a href="https://bitbucket.org/petsc/petsc/branch/knepley/fix-plex-interpolate-sf-new" class=""></a><a href="https://bitbucket.org/petsc/petsc/branch/knepley/fix-plex-interpolate-sf" class="">https://bitbucket.org/petsc/petsc/branch/knepley/fix-plex-interpolate-sf</a>-new and
it now needs to take that not-yet-implemented feature into account - I will update it and remind to me what cases are still failing.</div>
<div class=""><br class="">
</div>
<div class="">Thanks in advance for clarification!</div>
<div class=""><br class="">
</div>
<div class="">Vaclav</div>
</div>
</div>
</blockquote>
</div>
<br class="">
</div>
</body>
</html>