[Carpet] local reduction handles

[Erik Schnetter]

On Oct 22, 2007, at 11:04:05, Bela Szilagyi wrote:

> On Monday 22 October 2007 17:12:10 Erik Schnetter wrote:
>> If you only want to use local reduction operators, then you can just
>> activate thorn CactusBase/LocalReduce.  It is independent of Carpet.
>
> Loading thorn CactusBase/LocalReduce does not solve the problem.
>
> Apparently I hit a bug in thorn EllSOR:

[schnipp]

> That is, the handle of the local reduction operator is passed to  
> the global
> reduction function.

That is not an error; this is the way the new interface works.

> Replacing the call to  CCTK_LocalArrayReductionHandle with an  
> equivalent call
> to CCTK_ReductionHandle does not solve the issue.
>
>
> Apparently EllSOR is assuming the new interface.  Is there an easy  
> fix? (aside
> from using PUGH...)

No, there is unfortunately not.  The new interface is still not  
completely defined to work for mesh refinement, and then there is a  
bit of implementation work left over for LocalReduce and PUGH.   
Afterwards, Carpet and our physics codes need to be updated.  And  
finally, you would need to add support for mesh refinement to EllSOR.

-erik

-- 
Erik Schnetter <schnetter at cct.lsu.edu>

My email is as private as my paper mail.  I therefore support encrypting
and signing email messages.  Get my PGP key from www.keyserver.net.



-------------- next part --------------
A non-text attachment was scrubbed...
Name: PGP.sig
Type: application/pgp-signature
Size: 186 bytes
Desc: This is a digitally signed message part
Url : /archives/developers/attachments/20071022/3b10f645/attachment.pgp