[petsc-dev] Petsc "make test" have more failures for --with-openmp=1

Lawrence Mitchell wence at gmx.li
Fri Mar 19 10:16:57 CDT 2021

> On 19 Mar 2021, at 14:21, Jed Brown <jed at jedbrown.org> wrote:
> Notice how the permutations are contained within the vertices {0, ..., 8}, edges {9, ..., 24}, and cells {25, ..., 32}. I would like to get rid of that restriction, but you've said it would have significant non-local consequences so I haven't tried.

What we do, and I'm not sure if this is exactly what you want, is the following.

Use DMPlexGetOrdering to compute permutatations of points with RCM. As you say, this is stratified.

Then, we traverse cells in this RCM ordering and compute a permutation of the full plex chart greedily.


permutation = empty(pstart - pend)
i = 0
for cell in reordered_cells:
   points = plex.getclosure(cell)
   for point in points:
       if not seen(point):
           permutation[i] = point
           i += 1

Now when we create a section, we say

PetscSectionSetPermutation(sec, permutation)

And then when the section offsets are calculated, the points are visited and numbered in the order provided by the permutation.

This seems to work, here are the sparsity patterns for a Q3 mass matrix on an unstructured quad mesh (from gmsh). One with RCM ordering applied, the other without).

I guess the only issue is now one needs to always remember to traverse cells in the reordered_cells order (although if you call DMPlexPermute that is automatic?).



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20210319/17975584/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: gmsh-order.png
Type: image/png
Size: 52721 bytes
Desc: not available
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20210319/17975584/attachment-0002.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rcm-order.png
Type: image/png
Size: 23107 bytes
Desc: not available
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20210319/17975584/attachment-0003.png>

More information about the petsc-dev mailing list