dolfin team mailing list archive

[Garth Wells <668862@xxxxxxxxxxxxxxxxxx>]

This has been 'fixed' by using unordered_set in SparsityPattern. This is
a bit slower for standard cases, but much faster with global dofs.

Another issue is that PETScMatrix is very slow for large problems with
global dofs. EpetraMatrix is much faster in this case.

** Changed in: dolfin
       Status: Confirmed => Fix Committed

-- 
You received this bug notification because you are a member of DOLFIN
Team, which is subscribed to DOLFIN.
https://bugs.launchpad.net/bugs/668862

Title:
  Sparsity pattern computation dead slow with 'real' spaces

Status in DOLFIN:
  Fix Committed

Bug description:
  The computation of the sparsity pattern is dead slow when real spaces
  are used, i.e.,

      R = FunctionSpace(mesh, 'R', 0)

  To see the problem, compare the neumann-poisson and poisson demos with
  fine meshes (256x256)