[petsc-users] Invalid MAT_CLASSID

[Matthew Knepley]

On Mon, Oct 20, 2014 at 5:21 PM, Harshad Sahasrabudhe <hsahasra at purdue.edu>
wrote:

> Hi Matt,
>
> If I have 2 instances of PETSc, one which is initialized by LibMesh and
> the other which is initialized in my code, then would getting a Mat from
> one version to the other cause this issue? I know that this is not the
> right way to do it, I'm just trying to zero in on the problem.
>

No, this is a simple check that is common to any PETSc.

  Matt


> a) is not the problem, since I create and assemble the matrix using
> LibMesh API. b) is not it either, I ran a simple example and it works in
> that case. For c), I ran my code with Valgrind and didn't get any invalid
> write errors.
>
> Thanks,
> Harshad
>
> On Mon, Oct 20, 2014 at 4:19 PM, Matthew Knepley <knepley at gmail.com>
> wrote:
>
>> On Mon, Oct 20, 2014 at 3:17 PM, Harshad Sahasrabudhe <
>> hsahasra at purdue.edu> wrote:
>>
>>> I get the Mat object from libMesh::PetscMatrix<>::mat()
>>> <http://libmesh.github.io/libmesh/doxygen/classlibMesh_1_1PetscMatrix.html#aaca7a5afeb063716ee67fa11302c2aea>,
>>> and it's been created and assembled before I call MatGetType.
>>>
>>
>> But what you are getting is invalid, so either
>>
>>   a) it has not been created and assembled in LibMesh
>>
>>   b) the right thing is not getting passed back (LibMesh problem)
>>
>> or
>>
>>   c) you are overwriting memory and it manifests here (check with
>> valgrind)
>>
>> It is very easy to look in these structures using the debugger. We are
>> just checking an integer
>> in the header for the object that identifies it as a Mat.
>>
>>    Matt
>>
>>
>>> On Mon, Oct 20, 2014 at 4:12 PM, Matthew Knepley <knepley at gmail.com>
>>> wrote:
>>>
>>>> On Mon, Oct 20, 2014 at 3:07 PM, Harshad Sahasrabudhe <
>>>> hsahasra at purdue.edu> wrote:
>>>>
>>>>> Hi,
>>>>>
>>>>> I get the following error when I call MatGetType even though I'm
>>>>> calling the function with a Mat object.
>>>>>
>>>>> [0]PETSC ERROR: Invalid argument!
>>>>> [0]PETSC ERROR: Wrong type of object: Parameter # 1!
>>>>>
>>>>> What could be going wrong?
>>>>>
>>>>
>>>> You are not calling it with a valid Mat object. Did you call MatCreate?
>>>>
>>>>    Matt
>>>>
>>>>
>>>>> Thanks,
>>>>> Harshad
>>>>>
>>>>
>>>>
>>>>
>>>> --
>>>> 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
>>>>
>>>
>>>
>>
>>
>> --
>> 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
>>
>
>


-- 
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
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-users/attachments/20141020/9688c8dc/attachment-0001.html>