Hi Roberto, I've solved this before. Please see the solver options in tests/regression/test_moore_spence.py: https://github.com/firedrakeproject/firedrake/blob/master/tests/regression/t... The idea is to use a fieldsplit to take the Schur complement onto the R block; solve the top-left block with LU, and use GMRES on the Schur complement. Since the Schur complement is 1x1 it will solve in one iteration without a preconditioner. A warning: support for problems with variables in R is not great in firedrake. It's probably the one area (that I use) where firedrake's support is painfully weaker than FEniCS'. Patrick On 17/04/2020 04:22, Roberto Federico Ausas wrote:

Hello,

I need to solve a problem involving functions belonging to an R space.

For robustness, I need to use direct solvers for the linear part, but I'm getting the following:

ValueError: Monolithic matrix assembly is not supported for systems with R-space blocks.

Is there any way of doing this? Any help would be appreciated Regards

-- Roberto F. Ausas Instituto de Ciências Matemáticas e de Computação- ICMC/USP São Carlos, SP - Brasil _http://www.lmacc.icmc.usp.br/~ausas_

_______________________________________________ firedrake mailing list firedrake@imperial.ac.uk https://mailman.ic.ac.uk/mailman/listinfo/firedrake