Boyce, Don't mess with this bad advice. It is just making your live complicated. Use a fast MatAXPY() either write it yourself or wait for use.
Barry On Sep 27, 2010, at 11:30 AM, Boyce Griffith wrote: > > > On 9/27/10 10:47 AM, Hong Zhang wrote: >> Boyce, >> >>> The easiest way I can think of to handle this is to make P^T C P, then >>> extract the nonzero >>> structure with Mat with GetIJ() and create B. You would have to explicitly >>> put zeros in B >>> too or else AssemblyEnd() will shrink it. Maybe someone else can think of an >>> easier way. >> >> This is what I would suggest as well. > > It looks like MatGetRowIJ will do the trick for SeqAIJ matrices, but it > appears that similar functionality does not exist for MPIAIJ matrices. Is > there an approach that you would recommend for the parallel case? > > Thanks! > > -- Boyce
