[petsc-users] FETI-DP implementation and call sequence

[Jed Brown]

Satish Balay <balay at mcs.anl.gov> writes:

> Ok - looks like this error check in
> ISLocalToGlobalMappingRestoreBlockInfo() was added a couple of days
> back.
>
>>>>
> https://bitbucket.org/petsc/petsc/commits/cbc1caf078fb2bf42b82e0b5ac811b1101900405
>   PetscValidHeaderSpecific(mapping,IS_LTOGM_CLASSID,1);
> <<<
>
>  This is breaking PCISDestroy() - which is attempting to pass in a null for 'mapping'
>
>>>>>>>
>   if (pcis->ISLocalToGlobalMappingGetInfoWasCalled) {
>     ierr = ISLocalToGlobalMappingRestoreInfo((ISLocalToGlobalMapping)0,&(pcis->n_neigh),&(pcis->neigh),&(pcis->n_shared),&(pcis->shared));CHKERRQ(ierr);

Yuck.

>   }
> <<<<<<
>
> Commenting out the error check gets the code working.

I consider the error check to be correct; the code using it needs to be fixed.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 818 bytes
Desc: not available
URL: <http://lists.mcs.anl.gov/pipermail/petsc-users/attachments/20140905/c1d3963c/attachment.pgp>