[petsc-dev] SEGV using PETSc-GPU ViennaCL

[Dominic Meiser]

Can you keep petsc-dev at mcs.anl.gov cc'd please.

On 07/29/2014 03:33 PM, Lucas Ogura wrote:
> I don't have permission to send all the code, but executing 
> step-by-step I was able to isolate the part that give the error and 
> I'm sending it. The complete code its an iterative process that 
> operates over a CSR matrix.
> Running some more tests I realize that I'm not sure if my runtime 
> options are right:
> 1- I'm passing "-vec_type viennacl -mat_type aijviennacl". Is that 
> correct?
> 2- Shoud I add extra functions in the code besides the 
> MatSetFromOptions and VecSetFromOptions after the MatCreat and VecCreate?

A minimal example that demonstrates the bug is fine (in fact it's easier 
for us than just sending the whole application). However, could you send 
a code sample that can be compiled, please. The one you sent is missing 
includes, variable declarations, and setup.

Your command line options sound reasonable but it's not possible to say 
for sure without working code.

Thanks,
Dominic
>
>
>
> ------------------------------------------------------------------------
> Date: Tue, 29 Jul 2014 11:09:10 -0600
> From: dmeiser at txcorp.com
> To: lucasjdogura at hotmail.com; petsc-dev at mcs.anl.gov
> Subject: Re: [petsc-dev] SEGV using PETSc-GPU ViennaCL
>
> Can you send example code that reproduces this problem?
>
>
> On 07/29/2014 10:54 AM, Lucas Ogura wrote:
>
>     I am trying to use PETSc-dev for a while in a project at my
>     university. I need to use the GPU for matrix-vector operations and
>     the code that I'm working was already implemented using PETSc,
>     obviously I choose the PETSc ViennaCL-GPU solution. The original
>     code works perfectly, but when I use the GPU parameters, a SEGV
>     occurs in some test cases.
>     Valgrind says that is a problem with PetscMallocAlign at memalign
>     but, some times, it says that the problem is MatGetDiagonal.
>     I'm sending the configure.log because I don't know what are the
>     cause of the problem (hope the problem is the configuration), and
>     I would like to know if some similar problem were reported by
>     PETSc-ViennaCL users and PETSc-CUDA users too.
>
>     PS: My computer has a top AMD GPU and I'm not sure that I can find
>     a NVIDIA GPU computer to test my code, so OpenCL solutions are
>     better to me then CUDA.
>
>
>
> -- 
> Dominic Meiser
> Tech-X Corporation
> 5621 Arapahoe Avenue
> Boulder, CO 80303
> USA
> Telephone: 303-996-2036
> Fax: 303-448-7756
> www.txcorp.com  <http://www.txcorp.com>


-- 
Dominic Meiser
Tech-X Corporation
5621 Arapahoe Avenue
Boulder, CO 80303
USA
Telephone: 303-996-2036
Fax: 303-448-7756
www.txcorp.com

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20140729/31b9b139/attachment.html>