[Trilinos-Users] TraceMin-Davidson failure

[Hoemmen, Mark]

On 11/11/14, 12:00 PM, "trilinos-users-request at software.sandia.gov"
<trilinos-users-request at software.sandia.gov> wrote:
>>----- Original Message -----
>> From: maireni at cimne.upc.edu
>> To: "Alicia M Klinvex" <aklinvex at purdue.edu>
>> Sent: Monday, November 10, 2014 2:54:46 AM
>> Subject: Re: [Trilinos-Users] Returned mail: see transcript for details
>> 
>> Hi Alicia,
>> At first so sorry for reply you a lot time for questions.
>> I am using the example you defined in the list that use
>> TraceMin-Davidson to solve a generalized eigenvalue problem
>> using Tpetra (and reading file).
>> 
>> The thing is i can not yet run my eigen problem.
>> In fact i have this mistakes and no reported in google
>> exist.  Here it is.
>> 
>> Teuchos::ArrayRCP<unsigned
>> 
>>long>{ptr=0x46f9130,lowerOffset=0,upperOffset=8238,size=8239,node=0x4d6c7
>>c0,strong_count=2,weak_count=0}
>>   >> Allocating X_
>>   >> Allocating KX_
>>   >> Allocating MX_
>>   >> Allocating R_
>>   >> Allocating V_
>>   >> Allocating KV_
>>   >> Allocating MV_
>>   >> done allocating.
>> Computing MV
>> terminate called after throwing an instance of 'std::runtime_error'
>>    what():
>> /usr/local/trilinos-11.12.1/include/Tpetra_CrsMatrix_def.hpp:3517:
>> 
>> Throw number = 1
>> 
>> Throw test that evaluated to true: ! isFillComplete ()

Did you actually call fillComplete on the matrix?  In both Epetra and
Tpetra, it is necessary to call fillComplete on the matrix after
constructing it, before you may use it in a sparse matrix-vector
multiplication.

mfh