[petsc-users] Viewing DM Vector stored on Multiple GPUs

[Dominic Meiser]

Hi Ashwin,

I've been working on this issue the last couple of days and wanted to 
give you an update. First off, thanks for the test case. This has helped 
a lot in tracking down the issues. I've created a couple of unit tests 
based on your test case.

There were several independent bugs that prevented the VecView to work 
with -dm_vec_type cusp. I've fixed some of them in this branch:

https://bitbucket.org/dmeiser/petsc/branch/fix-cusp-dmda

Viewing a vector created from a DM should now be possible. For instance 
the test case you provided should work with this branch.

There are still some issues remaining. DMGlobalToLocalBegin/End and 
DMLocalToGlobalBegin/End are not working. I haven't been able to track 
these down yet. I'll return to this next week.

Cheers,
Dominic


On 06/24/2014 08:15 PM, Ashwin Srinath wrote:
> Hello, petsc-users
>
> I'm having trouble /viewing/ an mpicusp vector. Here's the simplest 
> case that reproduces the problem:
>
> int main(int argc, char** argv) {
>
>         PetscInitialize(&argc, &argv, NULL, NULL);
>
>         DM da;
>         Vec V;
>
>         DMDACreate2d(   PETSC_COMM_WORLD,
>                         DM_BOUNDARY_NONE, DM_BOUNDARY_NONE,
>                         DMDA_STENCIL_BOX,
>                         5, 5,
>                         PETSC_DECIDE, PETSC_DECIDE,
>                         1,
>                         1,
>                         NULL, NULL,
>                         &da);
>
>         DMCreateGlobalVector(da, &V);
>
>         VecSet(V, 1);
> VecView(V, PETSC_VIEWER_STDOUT_WORLD);
>
>         PetscFinalize();
>         return 0;
> }
>
> I get the error:
> [1]PETSC ERROR: Null argument, when expecting valid pointer
> [0]PETSC ERROR: Trying to copy from a null pointer
>
> I executed with the following command:
> mpiexec -n 2 ./main -dm_vec_type cusp -vec_type cusp
> Both GPUs are attached to two different processes.
>
> This program works fine for vecmpi vectors, i.e., -dm_vec_type mpi and 
> -vec_type mpi. Also, I don't get an error unless I try to /view/ the 
> vector. Can someone please point out what I'm doing wrong?
>
> Thanks for your time,
> Ashwin Srinath
>
>
>
>
>


-- 
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-users/attachments/20140718/4a61d14c/attachment.html>