<div dir="ltr"><div>My code runs OK on SUMMIT but ex56 does not. ex56 runs in serial but I get this segv in parallel. I also see these memcheck messages from PetscSFBcastAndOpBegin in my code and ex56. </div><div><br></div><div>I ran this in DDT and was able to get a stack trace and look at variables. THe segv is on sfbasic.c:148:</div><div><br></div><div>ierr = MPI_Startall_isend(bas->rootbuflen[PETSCSF_REMOTE],unit,bas->nrootreqs,rootreqs);CHKERRQ(ierr);<br></div><div><br></div><div>I did not see anything wrong with the variables here. The segv is on processor 1 of 2 (so the last process).</div><div><br></div><div>Any ideas?</div><div>Thanks,</div><div>Mark</div><div><br></div><div>#18 main (argc=<optimized out>, args=<optimized out>) at /ccs/home/adams/petsc-old/src/snes/tutorials/ex56.c:477 (at 0x0000000010006224)<br>#17 SNESSolve (snes=0x31064ac0, b=0x3241ac40, x=<optimized out>) at /autofs/nccs-svm1_home1/adams/petsc-old/src/snes/interface/snes.c:4515 (at 0x0000200000de3498)<br>#16 SNESSolve_NEWTONLS (snes=0x31064ac0) at /autofs/nccs-svm1_home1/adams/petsc-old/src/snes/impls/ls/ls.c:175 (at 0x0000200000e1b344)<br>#15 SNESComputeFunction (snes=0x31064ac0, x=0x312a3170, y=0x4a464bd0) at /autofs/nccs-svm1_home1/adams/petsc-old/src/snes/interface/snes.c:2378 (at 0x0000200000dd5024)<br>#14 SNESComputeFunction_DMLocal (snes=0x31064ac0, X=0x312a3170, F=0x4a464bd0, ctx=0x4a3a7ed0) at /autofs/nccs-svm1_home1/adams/petsc-old/src/snes/utils/dmlocalsnes.c:71 (at 0x0000200000dab058)<br>#13 DMGlobalToLocalBegin (dm=0x30fac020, g=0x312a3170, mode=<optimized out>, l=0x4a444c40) at /autofs/nccs-svm1_home1/adams/petsc-old/src/dm/interface/dm.c:2407 (at 0x0000200000b28a38)<br>#12 PetscSFBcastBegin (leafdata=0x200073c00a00, rootdata=0x200073a00000, unit=<optimized out>, sf=0x30f28980) at /ccs/home/adams/petsc-old/include/petscsf.h:189 (at 0x0000200000b28a38)<br>#11 PetscSFBcastAndOpBegin (sf=0x30f28980, unit=0x200021879ed0, rootdata=0x200073a00000, leafdata=0x200073c00a00, op=0x200021889c70) at /autofs/nccs-svm1_home1/adams/petsc-old/src/vec/is/sf/interface/sf.c:1337 (at 0x000020000045a230)<br>#10 PetscSFBcastAndOpBegin_Basic (sf=0x30f28980, unit=0x200021879ed0, rootmtype=<optimized out>, rootdata=0x200073a00000, leafmtype=<optimized out>, leafdata=0x200073c00a00, op=0x200021889c70) at /autofs/nccs-svm1_home1/adams/petsc-old/src/vec/is/sf/impls/basic/sfbasic.c:148 (at 0x00002000003b1d9c)<br>#9 PMPI_Startall () from /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/gcc-6.4.0/spectrum-mpi-10.3.1.2-20200121-awz2q5brde7wgdqqw4ugalrkukeub4eb/container/../lib/libmpi_ibm.so.3 (at 0x00002000217e3d98)<br>#8 mca_pml_pami_start () from /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/gcc-6.4.0/spectrum-mpi-10.3.1.2-20200121-awz2q5brde7wgdqqw4ugalrkukeub4eb/container/../lib/spectrum_mpi/mca_pml_pami.so (at 0x000020002555e6e0)<br>#7 pml_pami_persis_send_start () from /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/gcc-6.4.0/spectrum-mpi-10.3.1.2-20200121-awz2q5brde7wgdqqw4ugalrkukeub4eb/container/../lib/spectrum_mpi/mca_pml_pami.so (at 0x000020002555e29c)<br>#6 pml_pami_send () from /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/gcc-6.4.0/spectrum-mpi-10.3.1.2-20200121-awz2q5brde7wgdqqw4ugalrkukeub4eb/container/../lib/spectrum_mpi/mca_pml_pami.so (at 0x000020002555f69c)<br>#5 PAMI_Send_immediate () from /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/gcc-6.4.0/spectrum-mpi-10.3.1.2-20200121-awz2q5brde7wgdqqw4ugalrkukeub4eb/container/../lib/pami_port/libpami.so.3 (at 0x0000200025725814)<br>#4 PAMI::Protocol::Send::Eager<PAMI::Device::Shmem::PacketModel<PAMI::Device::ShmemDevice<PAMI::Fifo::WrapFifo<PAMI::Fifo::FifoPacket<64u, 4096u>, PAMI::Counter::IndirectBounded<PAMI::Atomic::NativeAtomic>, 256u>, PAMI::Counter::Indirect<PAMI::Counter::Native>, PAMI::Device::Shmem::CMAShaddr, 256u, 512u> >, PAMI::Device::IBV::PacketModel<PAMI::Device::IBV::Device, true> >::EagerImpl<(PAMI::Protocol::Send::configuration_t)5, true>::immediate(pami_send_immediate_t*) () from /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/gcc-6.4.0/spectrum-mpi-10.3.1.2-20200121-awz2q5brde7wgdqqw4ugalrkukeub4eb/container/../lib/pami_port/libpami.so.3 (at 0x00002000257e7bac)<br>#3 PAMI::Protocol::Send::EagerSimple<PAMI::Device::Shmem::PacketModel<PAMI::Device::ShmemDevice<PAMI::Fifo::WrapFifo<PAMI::Fifo::FifoPacket<64u, 4096u>, PAMI::Counter::IndirectBounded<PAMI::Atomic::NativeAtomic>, 256u>, PAMI::Counter::Indirect<PAMI::Counter::Native>, PAMI::Device::Shmem::CMAShaddr, 256u, 512u> >, (PAMI::Protocol::Send::configuration_t)5>::immediate_impl(pami_send_immediate_t*) () from /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/gcc-6.4.0/spectrum-mpi-10.3.1.2-20200121-awz2q5brde7wgdqqw4ugalrkukeub4eb/container/../lib/pami_port/libpami.so.3 (at 0x00002000257e7824)<br>#2 bool PAMI::Device::Interface::PacketModel<PAMI::Device::Shmem::PacketModel<PAMI::Device::ShmemDevice<PAMI::Fifo::WrapFifo<PAMI::Fifo::FifoPacket<64u, 4096u>, PAMI::Counter::IndirectBounded<PAMI::Atomic::NativeAtomic>, 256u>, PAMI::Counter::Indirect<PAMI::Counter::Native>, PAMI::Device::Shmem::CMAShaddr, 256u, 512u> > >::postPacket<2u>(unsigned long, unsigned long, void*, unsigned long, iovec (&) [2u]) () from /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/gcc-6.4.0/spectrum-mpi-10.3.1.2-20200121-awz2q5brde7wgdqqw4ugalrkukeub4eb/container/../lib/pami_port/libpami.so.3 (at 0x00002000257e6c18)<br>#1 PAMI::Device::Shmem::Packet<PAMI::Fifo::FifoPacket<64u, 4096u> >::writePayload(PAMI::Fifo::FifoPacket<64u, 4096u>&, iovec*, unsigned long) () from /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/gcc-6.4.0/spectrum-mpi-10.3.1.2-20200121-awz2q5brde7wgdqqw4ugalrkukeub4eb/container/../lib/pami_port/libpami.so.3 (at 0x00002000257c5a7c)<br>#0 __memcpy_power7 () from /lib64/libc.so.6 (at 0x00002000219eb84c)<br></div><div><br></div><br>========= Program hit CUDA_ERROR_INVALID_VALUE (error 1) due to "invalid argument" on CUDA API call to cuPointerGetAttribute. <br>=========     Saved host backtrace up to driver entry point at error<br>=========     Host Frame:/lib64/libcuda.so.1 (cuPointerGetAttribute + 0x178) [0x2d14a8]<br>=========     Host Frame:/ccs/home/adams/petsc-old/arch-summit-opt-gnu-cuda-omp/lib/libpetsc.so.3.013 (PetscSFBcastAndOpBegin + 0xd8) [0x3ba1a0]<br>=========     Host Frame:/ccs/home/adams/petsc-old/arch-summit-opt-gnu-cuda-omp/lib/libpetsc.so.3.013 (PetscSectionCreateGlobalSection + 0x948) [0x3c5b24]<br>=========     Host Frame:/ccs/home/adams/petsc-old/arch-summit-opt-gnu-cuda-omp/lib/libpetsc.so.3.013 (DMGetGlobalSection + 0x98) [0xa85e5c]<br>=========     Host Frame:/ccs/home/adams/petsc-old/arch-summit-opt-gnu-cuda-omp/lib/libpetsc.so.3.013 (DMPlexCreateRigidBody + 0xc4) [0x9a9ce4]<br>=========     Host Frame:./ex56 [0x5aec]<br>=========     Host Frame:/lib64/libc.so.6 [0x25200]<br>=========     Host Frame:/lib64/libc.so.6 (__libc_start_main + 0xc4) [0x253f4]<br>=========<br>========= Program hit CUDA_ERROR_INVALID_VALUE (error 1) due to "invalid argument" on CUDA API call to cuPointerGetAttribute. <br>=========     Saved host backtrace up to driver entry point at error<br>=========     Host Frame:/lib64/libcuda.so.1 (cuPointerGetAttribute + 0x178) [0x2d14a8]<br>=========     Host Frame:/ccs/home/adams/petsc-old/arch-summit-opt-gnu-cuda-omp/lib/libpetsc[1]PETSC ERROR: ------------------------------------------------------------------------<br>[1]PETSC ERROR: Caught signal number 11 SEGV: Segmentation Violation, probably memory access out of range<br>[1]PETSC ERROR: Try option -start_in_debugger or -on_error_attach_debugger<br>[1]PETSC ERROR: or see <a href="https://www.mcs.anl.gov/petsc/documentation/faq.html#valgrind">https://www.mcs.anl.gov/petsc/documentation/faq.html#valgrind</a><br>[1]PETSC ERROR: or try <a href="http://valgrind.org">http://valgrind.org</a> on GNU/linux and Apple Mac OS X to find memory corruption errors<br>[1]PETSC ERROR: configure using --with-debugging=yes, recompile, link, and run <br>[1]PETSC ERROR: to get more information on the crash.<br>[1]PETSC ERROR: --------------------- Error Message --------------------------------------------------------------<br>[1]PETSC ERROR: Signal received<br>[1]PETSC ERROR: See <a href="https://www.mcs.anl.gov/petsc/documentation/faq.html">https://www.mcs.anl.gov/petsc/documentation/faq.html</a> for trouble shooting.<br>[1]PETSC ERROR: Petsc Development GIT revision: v3.13.2-421-gab8fa13  GIT Date: 2020-06-22 13:25:32 -0400<br>[1]PETSC ERROR: ./ex56 on a arch-summit-opt-gnu-cuda-omp named h35n05 by adams Tue Jun 23 09:06:20 2020<br>[1]PETSC ERROR: Configure options --with-fc=0 --COPTFLAGS="-g -O -fPIC -DFP_DIM=2" --CXXOPTFLAGS="-g -O -fPIC " --FOPTFLAGS="-g -O -fPIC " --CUDAOPTFLAGS="-g -O -Xcompiler -rdynamic -lineinfo--with-ssl=0" --with-batch=0 --with-cxx=mpicxx --with-mpiexec="jsrun -g1" --with-cuda=1 --with-cudac=nvcc --download-p4est=1 --download-zlib --download-hdf5=1 --download-metis --download-parmetis --download-triangle --with-blaslapack-lib="-L/autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/gcc-6.4.0/netlib-lapack-3.8.0-wcabdyqhdi5rooxbkqa6x5d7hxyxwdkm/lib64 -lblas -llapack" --with-cc=mpicc --with-shared-libraries=1 --with-x=0 --with-64-bit-indices=0 --with-debugging=0 PETSC_ARCH=arch-summit-opt-gnu-cuda-omp --with-openmp=1 --with-threadsaftey=1 --with-log=1 PETSC_DIR=/ccs/home/adams/petsc-old --force<br>[1]PETSC ERROR: #1 User provided function() line 0 in  unknown file<br>--------------------------------------------------------------------------<br></div>