Thanks Lawrence.

> I tried. But all four versions work for me.
> ...
> If you built your own PETSc, can you say what commit you used?

I was indeed using my own petsc.  It was 3.8.4.

When I reinstalled firedrake using "unset PYTHONPATH" and "unset PETSC_DIR" before running the firedrake-install script ... I think this is the right way to get firedrake to install its own PETSC ... then all four version worked.

My concern with the working version is that firedrake chooses a petsc version from 2013 or so?  I think 3.4.2 is that old (http://ftp.mcs.anl.gov/pub/petsc/release-snapshots/).

(firedrake) ~/repos/mccarthy/stokes[master*]$ ./linflowstep.py -version
--------------------------------------------------------------------------
Petsc Development GIT revision: v3.4.2-20859-g7bfdb6492c  GIT Date: 2018-04-03 16:04:52 -0500
       The PETSc Team
    petsc-maint@mcs.anl.gov
 http://www.mcs.anl.gov/petsc/
See docs/changes/index.html for recent updates.
See docs/faq.html for problems.
See docs/manualpages/index.html for help. 
Libraries linked from /home/ed/firedrake/lib/python3.6/site-packages/petsc/lib
--------------------------------------------------------------------------
effective viscosity 1.46e+12 Pa s
reading mesh from glacier.msh ...
...

> Can you post the output of:
> firedrake-status?

Both with the reinstalled firedrake, and the old, it crashes; here is the reinstalled case:

(firedrake) ~/repos/mccarthy/stokes[master*]$ source ~/firedrake/bin/firedrake-status 
__requires__: command not found
bash: /home/ed/firedrake/bin/firedrake-status: line 4: syntax error near unexpected token `'pkg_resources''
bash: /home/ed/firedrake/bin/firedrake-status: line 4: `__import__('pkg_resources').require('firedrake==0.13.0+2004.gc8a194d9')'

Thanks for your help!

Ed


On Mon, Apr 23, 2018 at 2:31 PM, Ed Bueler <elbueler@alaska.edu> wrote:
Dear Firedrake --

My smooth progress with this awesome tool has hit a first barrier.

The attached linear stokes code works as is with the "VERSION 1" weak form.  Run simply as "python3 ./linflowstep.py".  The attached mesh file is read by default.

But with "grad()" replaced by "epsilon()", as recommended in the UFL manual*, which is VERSION 2, I get seg fault.  Slight modifications do or do not work according to mere value of the constant (see VERSION 3/4 ... weird).

Can you reproduce?  Any ideas?

I seek to use the strain rate tensor (= epsilon(u)) in a nonnewtonian stokes job, but this is blocking.

Thanks,

Ed


PPS  Any advice for using a debugger on firedrake codes?  I am used to debugging PETSc C codes using gdb, but that is not working for me yet.



--
Ed Bueler
Dept of Math and Stat and Geophysical Institute
University of Alaska Fairbanks
Fairbanks, AK 99775-6660
301C Chapman



--
Ed Bueler
Dept of Math and Stat and Geophysical Institute
University of Alaska Fairbanks
Fairbanks, AK 99775-6660
301C Chapman