[petsc-users] DA_XYZGHOSTED

[Barry Smith]

   Juha,

     You are correct. DA_XYZGHOSTED is only implemented for 1D arrays.

     It is straightforward, but requires careful understanding of the  
code, to make the changes for 2 and 3d.

     I'm sorry I cannot do it a the moment.

     If you, or someone else, could add this support, that would be  
great, we'd love to include the fixes in PETSc.

     Barry

On Mar 8, 2010, at 7:12 AM, Juha Jäykkä wrote:

> Hi!
>
> I was wondering what's the status of DA_XYZGHOSTED? It seems to be  
> implemented
> for 1D DAs only. Putting the boundary data at inside (i.e. non- 
> ghosted part)
> of the DA is not very nice since it necessitates separate handling  
> of the for-
> loops on the ranks that happen to be on the physical boundary of the  
> lattice.
> This makes the code unnecessarily complicated from users' point of  
> view.
> Having the ghosts available at the boundaries as well would be the  
> ideal place
> to put the boundary data into. I understand these ghost values are  
> never
> updated (if running with no periodicity), so just setting them in the
> beginning would be correct.
>
> Cheers,
> Juha
>
> -- 
> 		 -----------------------------------------------
> 		| Juha Jäykkä, juhaj at iki.fi			|
> 		| http://www.maths.leeds.ac.uk/~juhaj		|
> 		 -----------------------------------------------