I run in an interactive shell on SUMMIT and don't need --smpiargs=-gpu. I have been having a lot of problems getting the cuda test to be reproducible in the pipeline and SUMMIT was failing. -use_gpu_aware_mpi 0 fixed the test on SUMMIT, in that the test ran and it look OK, and I updated the test. Hoping it fixes the pipeline ex56/cuda diffs.
On Tue, Jun 23, 2020 at 11:07 AM Stefano Zampini <stefano.zamp...@gmail.com> wrote: > What did it? How are you running now to have everything working? Can you > post smpiargs and petsc options? > > Il giorno mar 23 giu 2020 alle ore 17:51 Mark Adams <mfad...@lbl.gov> ha > scritto: > >> >> >> On Tue, Jun 23, 2020 at 9:54 AM Jed Brown <j...@jedbrown.org> wrote: >> >>> Did you use --smpiargs=-gpu, and have you tried if the error is still >>> >> >> >> >>> there with -use_gpu_aware_mpi 0? >> >> >> That did it. Thanks, >> >> >>> I assume you're using -dm_vec_type cuda? >>> >>> Mark Adams <mfad...@lbl.gov> writes: >>> >>> > 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. >>> > >>> > I ran this in DDT and was able to get a stack trace and look at >>> variables. >>> > THe segv is on sfbasic.c:148: >>> > >>> > ierr = >>> > >>> MPI_Startall_isend(bas->rootbuflen[PETSCSF_REMOTE],unit,bas->nrootreqs,rootreqs);CHKERRQ(ierr); >>> > >>> > I did not see anything wrong with the variables here. The segv is on >>> > processor 1 of 2 (so the last process). >>> > >>> > Any ideas? >>> > Thanks, >>> > Mark >>> > >>> > #18 main (argc=<optimized out>, args=<optimized out>) at >>> > /ccs/home/adams/petsc-old/src/snes/tutorials/ex56.c:477 (at >>> > 0x0000000010006224) >>> > #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) >>> > #16 SNESSolve_NEWTONLS (snes=0x31064ac0) at >>> > /autofs/nccs-svm1_home1/adams/petsc-old/src/snes/impls/ls/ls.c:175 (at >>> > 0x0000200000e1b344) >>> > #15 SNESComputeFunction (snes=0x31064ac0, x=0x312a3170, y=0x4a464bd0) >>> at >>> > /autofs/nccs-svm1_home1/adams/petsc-old/src/snes/interface/snes.c:2378 >>> (at >>> > 0x0000200000dd5024) >>> > #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) >>> > #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) >>> > #12 PetscSFBcastBegin (leafdata=0x200073c00a00, >>> rootdata=0x200073a00000, >>> > unit=<optimized out>, sf=0x30f28980) at >>> > /ccs/home/adams/petsc-old/include/petscsf.h:189 (at 0x0000200000b28a38) >>> > #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) >>> > #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) >>> > #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) >>> > #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) >>> > #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) >>> > #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) >>> > #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) >>> > #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) >>> > #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) >>> > #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) >>> > #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) >>> > #0 __memcpy_power7 () from /lib64/libc.so.6 (at 0x00002000219eb84c) >>> > >>> > >>> > ========= Program hit CUDA_ERROR_INVALID_VALUE (error 1) due to >>> "invalid >>> > argument" on CUDA API call to cuPointerGetAttribute. >>> > ========= Saved host backtrace up to driver entry point at error >>> > ========= Host Frame:/lib64/libcuda.so.1 (cuPointerGetAttribute + >>> > 0x178) [0x2d14a8] >>> > ========= Host >>> > >>> Frame:/ccs/home/adams/petsc-old/arch-summit-opt-gnu-cuda-omp/lib/libpetsc.so.3.013 >>> > (PetscSFBcastAndOpBegin + 0xd8) [0x3ba1a0] >>> > ========= Host >>> > >>> Frame:/ccs/home/adams/petsc-old/arch-summit-opt-gnu-cuda-omp/lib/libpetsc.so.3.013 >>> > (PetscSectionCreateGlobalSection + 0x948) [0x3c5b24] >>> > ========= Host >>> > >>> Frame:/ccs/home/adams/petsc-old/arch-summit-opt-gnu-cuda-omp/lib/libpetsc.so.3.013 >>> > (DMGetGlobalSection + 0x98) [0xa85e5c] >>> > ========= Host >>> > >>> Frame:/ccs/home/adams/petsc-old/arch-summit-opt-gnu-cuda-omp/lib/libpetsc.so.3.013 >>> > (DMPlexCreateRigidBody + 0xc4) [0x9a9ce4] >>> > ========= Host Frame:./ex56 [0x5aec] >>> > ========= Host Frame:/lib64/libc.so.6 [0x25200] >>> > ========= Host Frame:/lib64/libc.so.6 (__libc_start_main + 0xc4) >>> > [0x253f4] >>> > ========= >>> > ========= Program hit CUDA_ERROR_INVALID_VALUE (error 1) due to >>> "invalid >>> > argument" on CUDA API call to cuPointerGetAttribute. >>> > ========= Saved host backtrace up to driver entry point at error >>> > ========= Host Frame:/lib64/libcuda.so.1 (cuPointerGetAttribute + >>> > 0x178) [0x2d14a8] >>> > ========= Host >>> > >>> Frame:/ccs/home/adams/petsc-old/arch-summit-opt-gnu-cuda-omp/lib/libpetsc[1]PETSC >>> > ERROR: >>> > >>> ------------------------------------------------------------------------ >>> > [1]PETSC ERROR: Caught signal number 11 SEGV: Segmentation Violation, >>> > probably memory access out of range >>> > [1]PETSC ERROR: Try option -start_in_debugger or >>> -on_error_attach_debugger >>> > [1]PETSC ERROR: or see >>> > https://www.mcs.anl.gov/petsc/documentation/faq.html#valgrind >>> > [1]PETSC ERROR: or try http://valgrind.org on GNU/linux and Apple Mac >>> OS X >>> > to find memory corruption errors >>> > [1]PETSC ERROR: configure using --with-debugging=yes, recompile, link, >>> and >>> > run >>> > [1]PETSC ERROR: to get more information on the crash. >>> > [1]PETSC ERROR: --------------------- Error Message >>> > -------------------------------------------------------------- >>> > [1]PETSC ERROR: Signal received >>> > [1]PETSC ERROR: See >>> https://www.mcs.anl.gov/petsc/documentation/faq.html >>> > for trouble shooting. >>> > [1]PETSC ERROR: Petsc Development GIT revision: v3.13.2-421-gab8fa13 >>> GIT >>> > Date: 2020-06-22 13:25:32 -0400 >>> > [1]PETSC ERROR: ./ex56 on a arch-summit-opt-gnu-cuda-omp named h35n05 >>> by >>> > adams Tue Jun 23 09:06:20 2020 >>> > [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 >>> > [1]PETSC ERROR: #1 User provided function() line 0 in unknown file >>> > >>> -------------------------------------------------------------------------- >>> >> > > -- > Stefano >