[petsc-dev] preallocation checks

[Barry Smith]

On Jan 20, 2012, at 11:02 AM, Jed Brown wrote:

> On Fri, Jan 20, 2012 at 10:52, Barry Smith <bsmith at mcs.anl.gov> wrote:
> But what about a MatConvert_whatever_SeqAIJ() that uses MatGetRow() to 
> determine counts for preallocation. We don't have such a beast?????
> 
> I think MatConvert_Basic should work that way, but you have to determine 
> whether values are in the diagonal block or not.
Not for seqaij you don't.   Come on, it is understandable we do a crummy job 
for a hard case but we should at least do a good job for an easy case.