<div dir="ltr">Mark,<div>   without cuda-memcheck, did the test run?<div><div><div><div dir="ltr" class="gmail_signature" data-smartmail="gmail_signature"><div dir="ltr">--Junchao Zhang</div></div></div><br></div></div></div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Sun, Sep 26, 2021 at 12:38 PM Mark Adams <<a href="mailto:mfadams@lbl.gov">mfadams@lbl.gov</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr"><div>FYI, I am getting this with cuda-memcheck on Summit with CUDA 11.0.3:</div><div><br></div><div>jsrun -n 48 -a 6 -c 6 -g 1 -r 6 --smpiargs -gpu cuda-memcheck ../ex13-cu -dm_plex_box_faces 4,6,12 -petscpartitioner_simple_node_grid 2,2,2 -dm_plex_box_upper 2,3,6 -petscpartitioner_simple_process_grid 2,3,6 -dm_refine 3 -dm_mat_type aijcusparse -dm_vec_type cuda -dm_view<br></div><div><br></div><div>This job runs with Kokkos and it runs with these 8 nodes with -dm_refine 2 instead of 3. And it runs with the 1 node version of this test.</div><div><br></div><div>Thanks,</div><div>Mark</div><div><br></div>[8]PETSC ERROR: --------------------- Error Message --------------------------------------------------------------<br>[8]PETSC ERROR: GPU error<br>[8]PETSC ERROR: cuda error 715 (cudaErrorIllegalInstruction) : an illegal instruction was encountered<br>[8]PETSC ERROR: See <a href="https://petsc.org/release/faq/" target="_blank">https://petsc.org/release/faq/</a> for trouble shooting.<br>[8]PETSC ERROR: Petsc Development GIT revision: v3.15.4-943-g83e1f11c26  GIT Date: 2021-09-25 18:33:40 -0400<br>[8]PETSC ERROR: ../ex13-cu on a arch-summit-opt-gnu-cuda named f10n17 by Unknown Sun Sep 26 13:29:03 2021<br>[8]PETSC ERROR: Configure options --with-fc=mpifort --with-cc=mpicc --with-cxx=mpiCC --CFLAGS="-fPIC -g -DLANDAU_DIM=2 -DLANDAU_MAX_SPECIES=10 -DLANDAU_MAX_Q=4" --CXXFLAGS="-fPIC -g -DLANDAU_DIM=2 -DLANDAU_MAX_SPECIES=10 -DLANDAU_MAX_Q=4" --CUDAFLAGS="-g -Xcompiler -DLANDAU_DIM=2 -DLANDAU_MAX_SPECIES=10 -DLANDAU_MAX_Q=4" --FCFLAGS="-fPIC -g" --COPTFLAGS=-O3 --CXXOPTFLAGS=-O3 --with-ssl=0 --with-batch=0 --with-mpiexec="jsrun -g1" --with-cuda=1 --with-cudac=nvcc --with-cuda-arch=70 --download-p4est=1 --download-zlib --with-blaslapack-lib="-L/sw/summit/spack-envs/base/opt/linux-rhel8-ppc64le/gcc-9.1.0/netlib-lapack-3.9.1-e6vxode53ghsjrop2dfwlq77s3dvkr7t/lib64 -lblas -llapack" --with-x=0 --with-64-bit-indices=0 --with-debugging=0 PETSC_ARCH=arch-summit-opt-gnu-cuda<br>[8]PETSC ERROR: #1 PetscSFLinkSyncStream_CUDA() at /gpfs/alpine/csc314/scratch/adams/petsc/src/vec/is/sf/impls/basic/cuda/<a href="http://sfcuda.cu:872" target="_blank">sfcuda.cu:872</a><br>[8]PETSC ERROR: #2 PetscSFLinkSyncStreamBeforeCallMPI() at /gpfs/alpine/csc314/scratch/adams/petsc/include/../src/vec/is/sf/impls/basic/sfpack.h:344<br>[8]PETSC ERROR: #3 PetscSFLinkStartRequests_MPI() at /gpfs/alpine/csc314/scratch/adams/petsc/src/vec/is/sf/impls/basic/sfmpi.c:40<br>[8]PETSC ERROR: #4 PetscSFLinkStartCommunication() at /gpfs/alpine/csc314/scratch/adams/petsc/include/../src/vec/is/sf/impls/basic/sfpack.h:267<br>[8]PETSC ERROR: #5 PetscSFBcastBegin_Basic() at /gpfs/alpine/csc314/scratch/adams/petsc/src/vec/is/sf/impls/basic/sfbasic.c:191<br>[8]PETSC ERROR: #6 PetscSFBcastWithMemTypeBegin() at /gpfs/alpine/csc314/scratch/adams/petsc/src/vec/is/sf/interface/sf.c:1493<br>[8]PETSC ERROR: #7 DMGlobalToLocalBegin() at /gpfs/alpine/csc314/scratch/adams/petsc/src/dm/interface/dm.c:2613<br>[8]PETSC ERROR: #8 SNESComputeJacobian_DMLocal() at /gpfs/alpine/csc314/scratch/adams/petsc/src/snes/utils/dmlocalsnes.c:119<br>[8]PETSC ERROR: #9 SNESComputeJacobian() at /gpfs/alpine/csc314/scratch/adams/petsc/src/snes/interface/snes.c:2824<br>[8]PETSC ERROR: #10 SNESSolve_KSPONLY() at /gpfs/alpine/csc314/scratch/adams/petsc/src/snes/impls/ksponly/ksponly.c:43<br>[8]PETSC ERROR: #11 SNESSolve() at /gpfs/alpine/csc314/scratch/adams/petsc/src/snes/interface/snes.c:4769<br>[8]PETSC ERROR: #12 main() at ex13.c:169<br></div>
</blockquote></div>