On Tue, 30 Jun 2015, Andrew Davis wrote:

> Thanks for the help and sorry for the slow reply; I've been using the 
> deprecated code for now.

That's fine; the only problem with the deprecated code is a race
condition when using threads, and libMesh only uses threads when
instructed to on the command line.

> I had to prepare for a conference presentation and didn't want to
> worry about updating or anything.

I just let your reply sit for a month while I was on vacation; your
excuse is better.

> I just did a git pull so I should have the most recent version of
> libmesh (0.9.4 I think).

In the future, if you're getting versions from git, the way to specify
the version is with the commit hash at the top of "git log -1".  If
you just say "the most recent version" then that underspecifies
things, since by the time some jerk answers your email a month later
the most recent version will have changed.

> I tried calling sub_point_locator() just after the mesh was
> constructed but still got the same warning.

Did you also change the later point_locator() call to
sub_point_locator()?
---
Roy

------------------------------------------------------------------------------
_______________________________________________
Libmesh-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/libmesh-users

Reply via email to