You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Matt says DMPlexFilter() does not create a submesh with the domain cells hanging off it. As a result, when we later try to project from the submesh to the domain mesh to set boundary conditions, the projection fails.
Matt says we should go back to using DMPlexCreateSubmesh() like we have been doing when we use vertices to specify the BCs.
I am reading Cubit and Gmsh files with boundaries specified by faces and when I try to set boundary values for Dirichlet BCs I get the error below:
The MMS tests are working, so there is something different between using a function for setting values as opposed to a field (projection).
This works:
This generates the error:
Reproduce
knepley/pylith
https://github.com/baagaard-usgs/pylith.git
feature-meshio-hdf5
The text was updated successfully, but these errors were encountered: