On Mon, Oct 13, 2014 at 12:59 PM, David Fuentes <[email protected]> wrote:
> Hi, > > when using the snes/ex12 example > https://bitbucket.org/petsc/petsc/src/f5476a2b443e0eedb22d31cb1b2d85b9b918f73f/src/snes/examples/tutorials/ex12.c?at=master > with > > -variable_coefficient field -petscfe_type opencl -mat_petscfe_type opencl > > and > > -dm_vec_type viennacl -dm_mat_type seqaijviennacl -mat_type seqaijviennacl > > will PetscFEIntegrateResidual_OpenCL transfer the residual vector directly > to the > viennacl datastructures on the GPU for the solve ? > is the output residual of PetscFEIntegrateResidual_OpenCL a vector on the > CPU ? > The answer is Yes, if I have not screwed anything up. I have not specifically checked, and this stuff is still fairly touchy. It should be easy to check by just forming one residual. There should be no entries in log_summary for the GPU communication. Karl, is this easy to check for you? I need to get everything reinstalled on a different machine. Thanks, Matt > thanks, > David > > > > -- 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
