[petsc-users] DMPlex ghost cells

[Matthew Knepley]

On Thu, Aug 3, 2017 at 10:52 PM, Adrian Croucher <a.croucher at auckland.ac.nz>
wrote:

> Would it also work if I just didn't use DMPlexGetHybridBounds() to
> differentiate between partition ghost cells and boundary ghost cells, but
> instead checked the value of the "ghost" DMLabel?
>
Yes.

   Matt

> It appears that DMPlexConstructGhostCells() labels boundary ghosts with
> the ghost label value 1, and partition ghosts with value 2 (in
> DMPlexShiftLabels_Internal()).
>
> If that is reliable, then it might save me having to make sure the
> boundary ghosts always come last.
>
> - Adrian
>
> On 04/08/17 15:11, Matthew Knepley wrote:
>
> On Thu, Aug 3, 2017 at 10:07 PM, Adrian Croucher <
> a.croucher at auckland.ac.nz> wrote:
>
>> I am adding cells onto a DMPlex (for dual porosity). I notice that
>> normally the ghost cells in a DMPlex are at the end of the cell list.
>>
>> Does it matter if the cells I add come after those ghost cells, or should
>> the ghosts always come last?
>>
>> It seems to me it probably shoudn't matter if the partition ghosts
>> (created by DMPlexDistribute()) are not at the end. In my cell loops I just
>> check the value of the ghost DMLabel and skip them if necessary.
>>
>> But perhaps the boundary ghosts (created by DMPlexConstructGhostCells())
>> need to come last? Otherwise the output from DMPlexGetHybridBounds() (to
>> get the bounds on 'interior' cells, i.e. non-boundary-ghosts) probably
>> wouldn't make much sense?
>
>
> Yes, the only reason to have them last is to have GetHybridBounds() work
> for that case.
>
>   Thanks,
>
>      Matt
>
>
>>
>> - Adrian
>>
>> --
>> Dr Adrian Croucher
>> Senior Research Fellow
>> Department of Engineering Science
>> University of Auckland, New Zealand
>> email: a.croucher at auckland.ac.nz
>> tel: +64 (0)9 923 4611
>>
>>
>
>
> --
> What most experimenters take for granted before they begin their
> experiments is infinitely more interesting than any results to which their
> experiments lead.
> -- Norbert Wiener
>
> http://www.caam.rice.edu/~mk51/
>
>
> --
> Dr Adrian Croucher
> Senior Research Fellow
> Department of Engineering Science
> University of Auckland, New Zealand
> email: a.croucher at auckland.ac.nz
> tel: +64 (0)9 923 4611 <+64%209-923%204611>
>
>


-- 
What most experimenters take for granted before they begin their
experiments is infinitely more interesting than any results to which their
experiments lead.
-- Norbert Wiener

http://www.caam.rice.edu/~mk51/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-users/attachments/20170804/2f2fde23/attachment-0001.html>