- changed title to NonlinearDiscreteProblem should use system assembler if symmetric==True
- edited description
NonlinearDiscreteProblem should use system assembler if symmetric==True
Issue #225
resolved
Result of the long discussion is the revelation of this issue:
NonlinearVariationalSolver::NonlinearDiscreteProblem::J/F
uses normal Assembler
breaking a symmetry even if parameter NonlinearVariationalSolver.parameters['symmetric'] == true
. In fact, this parameter is not used anywhere.
Comments (9)
-
reporter -
- changed milestone to 1.5
-
reporter I wouldn't hurry with this for the same reason as https://bitbucket.org/fenics-project/dolfin/issue/167/cleanup-of-linear-solver-logic-needed#comment-14043406
-
- changed milestone to 1.6
-
- changed milestone to 1.7
-
reporter Issue
#621was marked as a duplicate of this issue. -
- changed status to resolved
Fixed in 2c293c5 by using symmetric assembler in all cases. Option 'symmetric' yet to be removed.
-
reporter Option 'symmetric' yet to be removed.
Not necessarily. Cholesky/MINRES can be invoked rather than LU/GMRES.
-
- removed milestone
Removing milestone: 1.7 (automated comment)
- Log in to comment
symmetry -> symmetric