Colin, So I didn't read the paper and therefore don't know exactly what you're trying to implement, but the PETSc folks have some examples on how to precondition the Stokes equation using command-line options starting around slide 129: http://www.mcs.anl.gov/petsc/documentation/tutorials/BuffaloTutorial.pdf Perhaps one of their examples closely resembles what you're looking for? Justin On Sat, Aug 20, 2016 at 11:12 AM, Colin Cotter <colin.cotter@imperial.ac.uk> wrote:

Dear all, I'd very much like to try the preconditioner in Mardal, Kent-Andre, and Ragnar Winther. "Uniform preconditioners for the time dependent Stokes problem." *Numerische Mathematik* 98.2 (2004): 305-327.

This is an additive preconditioner that involves summing the result of two different preconditioning systems. What is the best path to trying this right now? Is it the new matrix-free stuff?

cheers --cjc

-- http://www.imperial.ac.uk/people/colin.cotter

www.cambridge.org/9781107663916

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