Hi Timo,

thanks for replying. I think I didn't describe my problem correctly. The 4 
blocks have sizes: 1815 x 1815, 1815 x 330, 330 x 1815, 330 x 330. Such a 
division results from having 1815 displacement DoFs and 330 electric 
potential DoFs in the eigenvalue problem. Only the block 1815 x 1815 of the 
mass matrix has non-zero entries and there are 330 zeros in the diagonal of 
the mass matrix. I would like to condense the system to get rid of the zero 
diagonal entries in the mass matrix using the operations on the blocks. 

Since the total number of the DoFs in the system is 2145, all the 
IndexSets, that are passed to the reinit() function of the 
PETSc::MPI::BlockSparseMatrix are of size 2145. 

This results in failed assertion, since: rows[0].size() = 2145, 
rows[1].size() = 2145, but bdsp.block(0,0).n_rows() = 1815, 
bdsp.block(1,1).n_rows() = 330 etc...

Sorry, I cannot post the screenshots of these values since I am on holidays 
starting from today, I remember them from previous days as I checked them 
already at that time.

I tried to explicitly set the size of the IndexSets, i.e. for the IndexSet 
corresponding to electric potential DoFs that would be 330. Then, I cannot 
add indices higher than 330 to this IndexSet due to its internal structure.

Is there a way to overcome that or is the assertion itself wrong?

Best regards,
Gabriel

-- 
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/b7505a7d-f20c-4e20-b4dd-c745b8f1dccdn%40googlegroups.com.

Reply via email to