Kien, > I looked through the documentation in Deal.II before. > I think it could be done, too. It would be easier if there is an example > to use the PETSc parameters for MATSOLVERMUMPS package in deal.II. > Currently, I don't know where to put it in my codes. >
As Bruno has said this should work using the command line. When you run the program you could e.g. append some option like -mat_mumps_icntl_4 3 to set ICNTL(4) to 3. Not sure, but maybe you could do some trickery to modify argc and argv if you want to hard code (though this is certainly not a nice solution). Anyway, the SparseDirectMUMPS class is rather compact - so if you really want to customize, it shouldn't be too hard to write your own class using SparseDirectMUMPS as a sample. But it really depends on what you want to achieve. If you only want to select the factorization method, you can just use set_symmetric_mode(false/true) to switch between "LU" and "Cholesky" (which actually is LDL^T). Regards, Sebastian -- The deal.II project is located at http://www.dealii.org/ For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en --- You received this message because you are subscribed to the Google Groups "deal.II User Group" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/dealii/13a929d2-45d0-4388-b441-55e149ac9e6e%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
