Re: [deal.II] Re: Installation of deal.II with CUDA
Hi Bruno, > If I could get involved in this development, I would be helpful to > optimize the memory transfer between GPU and CPU by testing different CUDA > C programming schemes like this: > https://devblogs.nvidia.com/how-overlap-data-transfers-cuda-cc/ > That would be nice. Playing with streams is not something that we have > time to do in the foreseeable future. So if you are interested in > doing it, that would be a great!. > I am interested in testing the overlapping and different schemes developed you and others in deal.II. I have one question: if I want to install the latest "develop" version, I just simply type this: git clone https://github.com/dealii/dealii, right? If not, which command I need to type? Thanks! Sincerely, Chih-Che -- The deal.II project is located at http://www.dealii.org/ For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en --- You received this message because you are subscribed to the Google Groups "deal.II User Group" group. To unsubscribe from this group and stop receiving emails from it, send an email to dealii+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [deal.II] Re: Installation of deal.II with CUDA
Daniel, Using MPI and CUDA together is pretty much work in progress. On the > developer branch, LinearAlgebra::distributed::Vector supports > storing its elements on a CUDA device, but the values are copied to the > CPU for MPI communication. > The pull request https://github.com/dealii/dealii/pull/7303 allows to use > CUDA-aware MPI instead but the memory transfer needs to be optimized. > If I could get involved in this development, I would be helpful to optimize the memory transfer between GPU and CPU by testing different CUDA C programming schemes like this: https://devblogs.nvidia.com/how-overlap-data-transfers-cuda-cc/ In fact, as a principal investigator at Taiwan, I am qualified to submit jobs into Taiwania, a recently established super computing machine established in 2017, equipped with NVIDIA Tesla P100 SXM2. Moreover, sometime next year, we will have another super computing machine available to use (Taiwania 2), which now ranks 20th in the world with NVIDIA Tesla V100 SXM2. https://www.top500.org/system/179590 Sincerely, Chih-Che -- The deal.II project is located at http://www.dealii.org/ For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en --- You received this message because you are subscribed to the Google Groups "deal.II User Group" group. To unsubscribe from this group and stop receiving emails from it, send an email to dealii+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [deal.II] Re: Installation of deal.II with CUDA
Bruno, You are right. With installing gcc 5 and module load it, deal.II can be successfully installed with CUDA in the GPU machine I am using. Thanks! Right now, I would like to know this: is there any existing deal.II code example I could use to run on a machine with CUDA-aware MPI? Thanks! Sincerely, Chih-Che On Thu, Nov 8, 2018 at 10:58 PM Bruno Turcksin wrote: > Chih-Che, > > Le jeu. 8 nov. 2018 à 09:49, Chih-Che Chueh a écrit : > > -- The CUDA compiler identification is unknown > > -- Check for working CUDA compiler: /pkg/cuda/8.0.61/bin/nvcc > > -- Check for working CUDA compiler: /pkg/cuda/8.0.61/bin/nvcc -- broken > > CMake Error at /usr/share/cmake3/Modules/CMakeTestCUDACompiler.cmake:46 > (message): > > The CUDA compiler > > > > "/pkg/cuda/8.0.61/bin/nvcc" > > > > is not able to compile a simple test program. > > > > It fails with the following output: > > > > Change Dir: /work1/chueh5851a/CUDA/dealii/build/CMakeFiles/CMakeTmp > > > > Run Build Command:"/usr/bin/gmake" "cmTC_67f8a/fast" > > /usr/bin/gmake -f CMakeFiles/cmTC_67f8a.dir/build.make > CMakeFiles/cmTC_67f8a.dir/build > > gmake[1]: Entering directory > `/work1/chueh5851a/CUDA/dealii/build/CMakeFiles/CMakeTmp' > > Building CUDA object CMakeFiles/cmTC_67f8a.dir/main.cu.o > > /pkg/cuda/8.0.61/bin/nvcc -x cu -c > /work1/chueh5851a/CUDA/dealii/build/CMakeFiles/CMakeTmp/main.cu -o > CMakeFiles/cmTC_67f8a.dir/main.cu.o > > nvcc warning : The 'compute_20', 'sm_20', and 'sm_21' architectures > are deprecated, and may be removed in a future release (Use > -Wno-deprecated-gpu-targets to suppress warning). > > In file included from > /pkg/cuda/8.0.61/bin/..//include/cuda_runtime.h:78:0, > > from :0: > > /pkg/cuda/8.0.61/bin/..//include/host_config.h:119:2: error: #error > -- unsupported GNU version! gcc versions later than 5 are not supported! > > #error -- unsupported GNU version! gcc versions later than 5 are > not supported! > > ^ > > gmake[1]: *** [CMakeFiles/cmTC_67f8a.dir/main.cu.o] Error 1 > > gmake[1]: Leaving directory > `/work1/chueh5851a/CUDA/dealii/build/CMakeFiles/CMakeTmp' > > gmake: *** [cmTC_67f8a/fast] Error 2 > This is because of a change we just did here > <https://github.com/dealii/dealii/pull/7370>. You will need to use GCC 5 > instead of GCC 6. > > Best, > > Bruno > > -- > The deal.II project is located at http://www.dealii.org/ > For mailing list/forum options, see > https://groups.google.com/d/forum/dealii?hl=en > --- > You received this message because you are subscribed to the Google Groups > "deal.II User Group" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to dealii+unsubscr...@googlegroups.com. > For more options, visit https://groups.google.com/d/optout. > -- The deal.II project is located at http://www.dealii.org/ For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en --- You received this message because you are subscribed to the Google Groups "deal.II User Group" group. To unsubscribe from this group and stop receiving emails from it, send an email to dealii+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [deal.II] Re: Installation of deal.II with CUDA
Bruno > See the attached file for your reference and checking. > The testsuite only runs for CUDA 8.0 and 9.0 can you try one of this > version of CUDA and see if that helps? > When I am using cuda/8.0.61, I get the following error after I do cmake. [chueh5851a@glogin1 build]$ module list Currently Loaded Modulefiles: 1) gcc/6.3.03) blas/gcc/64/3.7.0 5) scalapack/openmpi/gcc/64/2.0.2 2) mpi/openmpi-3.0.0/gcc630 4) lapack/gcc/64/3.7.0 6) cuda/8.0.61 -- Include /work1/chueh5851a/CUDA/dealii/cmake/configure/configure_1_cuda.cmake -- Found CUDA: /pkg/cuda/8.0.61 (found version "8.0") -- Configured to use CUDA installation at /pkg/cuda/8.0.61 -- CUDA_VERSION: 8.0 -- CUDA_LIBRARIES: /pkg/cuda/8.0.61/lib64/libcudart.so;/pkg/cuda/8.0.61/lib64/libcusparse.so;/pkg/cuda/8.0.61/lib64/libcusolver.so -- CUDA_INCLUDE_DIRS: /pkg/cuda/8.0.61/include -- CUDA_USER_INCLUDE_DIRS: /pkg/cuda/8.0.61/include -- Found CUDA -- Found OpenMP_C: -fopenmp (found version "4.5") -- Found OpenMP_CXX: -fopenmp (found version "4.5") -- Found OpenMP_Fortran: -fopenmp (found version "4.0") -- Found OpenMP: TRUE (found version "4.5") -- The CUDA compiler identification is unknown -- Check for working CUDA compiler: /pkg/cuda/8.0.61/bin/nvcc -- Check for working CUDA compiler: /pkg/cuda/8.0.61/bin/nvcc -- broken CMake Error at /usr/share/cmake3/Modules/CMakeTestCUDACompiler.cmake:46 (message): The CUDA compiler "/pkg/cuda/8.0.61/bin/nvcc" is not able to compile a simple test program. It fails with the following output: Change Dir: /work1/chueh5851a/CUDA/dealii/build/CMakeFiles/CMakeTmp Run Build Command:"/usr/bin/gmake" "cmTC_67f8a/fast" /usr/bin/gmake -f CMakeFiles/cmTC_67f8a.dir/build.make CMakeFiles/cmTC_67f8a.dir/build gmake[1]: Entering directory `/work1/chueh5851a/CUDA/dealii/build/CMakeFiles/CMakeTmp' Building CUDA object CMakeFiles/cmTC_67f8a.dir/main.cu.o /pkg/cuda/8.0.61/bin/nvcc -x cu -c /work1/chueh5851a/CUDA/dealii/build/CMakeFiles/CMakeTmp/main.cu -o CMakeFiles/cmTC_67f8a.dir/main.cu.o nvcc warning : The 'compute_20', 'sm_20', and 'sm_21' architectures are deprecated, and may be removed in a future release (Use -Wno-deprecated-gpu-targets to suppress warning). In file included from /pkg/cuda/8.0.61/bin/..//include/cuda_runtime.h:78:0, from :0: /pkg/cuda/8.0.61/bin/..//include/host_config.h:119:2: error: #error -- unsupported GNU version! gcc versions later than 5 are not supported! #error -- unsupported GNU version! gcc versions later than 5 are not supported! ^ gmake[1]: *** [CMakeFiles/cmTC_67f8a.dir/main.cu.o] Error 1 gmake[1]: Leaving directory `/work1/chueh5851a/CUDA/dealii/build/CMakeFiles/CMakeTmp' gmake: *** [cmTC_67f8a/fast] Error 2 CMake will not be able to correctly generate this project. Call Stack (most recent call first): cmake/configure/configure_1_cuda.cmake:124 (ENABLE_LANGUAGE) build/CMakeFiles/CMakeTmp/evaluate_expression.tmp:1 (FEATURE_CUDA_CONFIGURE_EXTERNAL) cmake/macros/macro_evaluate_expression.cmake:30 (INCLUDE) cmake/macros/macro_configure_feature.cmake:244 (EVALUATE_EXPRESSION) cmake/configure/configure_1_cuda.cmake:164 (CONFIGURE_FEATURE) cmake/macros/macro_verbose_include.cmake:19 (INCLUDE) CMakeLists.txt:124 (VERBOSE_INCLUDE) -- Configuring incomplete, errors occurred! See also "/work1/chueh5851a/CUDA/dealii/build/CMakeFiles/CMakeOutput.log". See also "/work1/chueh5851a/CUDA/dealii/build/CMakeFiles/CMakeError.log". Sincerely, Chih-Che -- The deal.II project is located at http://www.dealii.org/ For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en --- You received this message because you are subscribed to the Google Groups "deal.II User Group" group. To unsubscribe from this group and stop receiving emails from it, send an email to dealii+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [deal.II] Re: Installation of deal.II with CUDA
Bruno, > make[1]: *** > [source/matrix_free/CMakeFiles/obj_matrix_free_debug.dir/all] Error 2 > > make[1]: *** Waiting for unfinished jobs > It looks like the first error is before what you showed. Also can you > post the detailed.log file. > See the attached file for your reference and checking. Sincerely, Chih-Che -- The deal.II project is located at http://www.dealii.org/ For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en --- You received this message because you are subscribed to the Google Groups "deal.II User Group" group. To unsubscribe from this group and stop receiving emails from it, send an email to dealii+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout. detailed.log Description: Binary data
[deal.II] Installation of deal.II with CUDA
Hi all, I am trying installing dealII with CUDA in a GPG machine by following this: https://www.dealii.org/9.0.0/external-libs/cuda.html But I get this error in terms of matrix_tools.cc.o. Is this because I don't turn off Trilinos stuff while cmaking deal.II for successful installation with CUDA? /usr/bin/cmake3 \ -DCMAKE_INSTALL_PREFIX=/work1/chueh5851a/software/dealii \ -DP4EST_DIR=/work1/chueh5851a/software/p4est-2.0/FAST \ -DTRILINOS_DIR=/work1/chueh5851a/software/Trilinos \ -D DEAL_II_WITH_MPI:BOOL=ON \ -D DEAL_II_WITH_THREADS:BOOL=ON \ -D DEAL_II_WITH_UMFPACK:BOOL=ON \ -D DEAL_II_WITH_TRILINOS:BOOL=ON \ -D DEAL_II_WITH_CUDA=ON \ -D DEAL_II_WITH_CXX14=OFF \ -D DEAL_II_WITH_PETSC:BOOL=OFF \ -D DEAL_II_WITH_SLEPC:BOOL=OFF \ `pwd`/.. ===make Errors ... [ 56%] Building CXX object source/numerics/CMakeFiles/obj_numerics_debug.dir/matrix_tools.cc.o [ 56%] Building CXX object source/lac/CMakeFiles/obj_lac_debug.dir/matrix_out.cc.o /work1/chueh5851a/software/Trilinos/include/impl/Kokkos_ConcurrentBitset.hpp(146): error: namespace "Kokkos" has no member "atomic_fetch_add" /work1/chueh5851a/software/Trilinos/include/impl/Kokkos_ConcurrentBitset.hpp(154): error: namespace "Kokkos" has no member "atomic_fetch_add" /work1/chueh5851a/software/Trilinos/include/impl/Kokkos_ConcurrentBitset.hpp(169): error: namespace "Kokkos" has no member "atomic_fetch_or" /work1/chueh5851a/software/Trilinos/include/impl/Kokkos_ConcurrentBitset.hpp(237): error: namespace "Kokkos" has no member "atomic_fetch_add" /work1/chueh5851a/software/Trilinos/include/impl/Kokkos_ConcurrentBitset.hpp(245): error: namespace "Kokkos" has no member "atomic_fetch_add" /work1/chueh5851a/software/Trilinos/include/impl/Kokkos_ConcurrentBitset.hpp(260): error: namespace "Kokkos" has no member "atomic_fetch_or" /work1/chueh5851a/software/Trilinos/include/impl/Kokkos_ConcurrentBitset.hpp(304): error: namespace "Kokkos" has no member "atomic_fetch_and" /work1/chueh5851a/software/Trilinos/include/impl/Kokkos_ConcurrentBitset.hpp(314): error: namespace "Kokkos" has no member "atomic_fetch_add" /work1/chueh5851a/software/Trilinos/include/impl/Kokkos_ConcurrentBitset.hpp(338): error: namespace "Kokkos" has no member "atomic_fetch_or" /work1/chueh5851a/software/Trilinos/include/impl/Kokkos_ConcurrentBitset.hpp(348): error: namespace "Kokkos" has no member "atomic_fetch_add" /work1/chueh5851a/software/Trilinos/include/impl/Kokkos_TaskQueue.hpp(226): error: namespace "Kokkos" has no member "atomic_exchange" /work1/chueh5851a/software/Trilinos/include/impl/Kokkos_HostBarrier.hpp(110): error: namespace "Kokkos" has no member "atomic_fetch_add" /work1/chueh5851a/software/Trilinos/include/impl/Kokkos_HostBarrier.hpp(113): error: namespace "Kokkos" has no member "atomic_fetch_add" /work1/chueh5851a/software/Trilinos/include/impl/Kokkos_HostBarrier.hpp(130): error: namespace "Kokkos" has no member "atomic_fetch_sub" /work1/chueh5851a/software/Trilinos/include/impl/Kokkos_HostBarrier.hpp(131): error: namespace "Kokkos" has no member "atomic_fetch_add" /work1/chueh5851a/software/Trilinos/include/impl/Kokkos_HostBarrier.hpp(239): error: namespace "Kokkos" has no member "atomic_fetch_add" /work1/chueh5851a/software/Trilinos/include/Sacado_DynamicArrayTraits.hpp(93): error: namespace "Kokkos" has no member "Cuda" /work1/chueh5851a/software/Trilinos/include/Sacado_DynamicArrayTraits.hpp(94): error: namespace "Kokkos" has no member "Cuda" /work1/chueh5851a/software/Trilinos/include/Sacado_DynamicArrayTraits.hpp(98): error: namespace "Kokkos" has no member "Cuda" /work1/chueh5851a/software/Trilinos/include/Sacado_DynamicArrayTraits.hpp(102): error: namespace "Kokkos" has no member "Cuda" /work1/chueh5851a/software/Trilinos/include/Sacado_DynamicArrayTraits.hpp(113): error: namespace "Kokkos" has no member "Cuda" /work1/chueh5851a/software/Trilinos/include/Sacado_DynamicArrayTraits.hpp(120): error: namespace "Kokkos" has no member "Cuda" /work1/chueh5851a/software/Trilinos/include/Kokkos_MemoryPool.hpp(112): error: a class or namespace qualified name is required detected during instantiation of class "Kokkos::MemoryPool [with DeviceType=]" /work1/chueh5851a/software/Trilinos/include/Sacado_DynamicArrayTraits.hpp(122): here /work1/chueh5851a/software/Trilinos/include/Kokkos_MemoryPool.hpp(112): error: global-scope qualifier (leading "::") is not allowed detected during instantiation of class "Kokkos::MemoryPool [with DeviceType=]" /work1/chueh5851a/software/Trilinos/include/Sacado_DynamicArrayTraits.hpp(122): here /work1/chueh5851a/software/Trilinos/include/Kokkos_MemoryPool.hpp(112): error: expected a ";" detected during instantiation of class "Kokkos::MemoryPool [with DeviceType=]" /work1/chueh5851a/software/Trilinos/include/Sacado_DynamicArrayTraits.hpp(122): here /work1/chueh5851a/software/Trilinos/include/Kokkos_MemoryPool.hpp(116): error: identifier
Re: [deal.II] Dealii installation error
Hi Wolfgang, I am happy to say that I can run both step-32.debug and step-32.release successfully in parallel. Basically, I re-install p4est again by following the steps: ./configure --enable-shared --enable-mpi --prefix=/home/chueh5851a/software/p4est-2.0/FAST make -j12 &> OUT.log make install cp -pr /home/chueh5851a/software/p4est-2.0/FAST /home/chueh5851a/software/p4est-2.0/DEBUG I guess that the problem was caused by the last step shown above: cp -pr /home/chueh5851a/software/p4est-2.0/FAST /home/chueh5851a/software/p4est-2.0/DEBUG At that time, I was typing cp -rf /home/chueh5851a/software/p4est-2.0/FAST/ /home/chueh5851a/software/p4est-2.0/DEBUG/ I am glad to say that I am able to use deal.II in Taiwan's largest supercomputing machines in Taiwan. :-) [chueh5851a@clogin2 step-32]$ mpirun -np 4 ./step-32.debug step-32.prm Number of active cells: 12,288 (on 6 levels) Number of degrees of freedom: 186,624 (99,840+36,864+49,920) Timestep 0: t=0 years Rebuilding Stokes preconditioner... Solving Stokes system... 41 iterations. Maximal velocity: 60.4935 cm/year Time step: 18166.9 years 17 CG iterations for temperature Temperature range: 973 4273.16 Number of active cells: 15,843 (on 7 levels) Number of degrees of freedom: 251,589 (136,040+47,529+68,020) Timestep 0: t=0 years Rebuilding Stokes preconditioner... Solving Stokes system... 50 iterations. Maximal velocity: 60.2831 cm/year Time step: 10557.8 years 19 CG iterations for temperature Temperature range: 973 4273.16 [chueh5851a@clogin2 step-32]$ mpirun -np 4 ./step-32.release step-32.prm Number of active cells: 12,288 (on 6 levels) Number of degrees of freedom: 186,624 (99,840+36,864+49,920) Timestep 0: t=0 years Rebuilding Stokes preconditioner... Solving Stokes system... 41 iterations. Maximal velocity: 60.4935 cm/year Time step: 18166.9 years 17 CG iterations for temperature Temperature range: 973 4273.16 Number of active cells: 15,843 (on 7 levels) Number of degrees of freedom: 251,589 (136,040+47,529+68,020) Timestep 0: t=0 years Rebuilding Stokes preconditioner... Solving Stokes system... 50 iterations. Maximal velocity: 60.2831 cm/year Time step: 10557.8 years 19 CG iterations for temperature Temperature range: 973 4273.16 Cheers, Chih-Che -- The deal.II project is located at http://www.dealii.org/ For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en --- You received this message because you are subscribed to the Google Groups "deal.II User Group" group. To unsubscribe from this group and stop receiving emails from it, send an email to dealii+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [deal.II] Dealii installation error
Hi Wolfgang, I don't know what is happening, but I see that all of the cases you show > run > the tutorial programs in release mode. Can you run them in debug mode and > see > whether you can get a better error message this way? > I will try this later after I figure out this warning happening at the end of configuring p4est (see the red warning messages), as this warning might trigger the unknown error I face now in our machine. ... checking for dgecon_... no checking for dgecon_ in -llapack... yes checking for LAPACK by linking w/ BLAS_FLIBS but w/o FLIBS... successful checking for getopt... successful checking for obstack... successful checking for library containing pthread_create... none required checking for POSIX threads... not used checking for library containing omp_get_thread_num... -lgomp checking for OpenMP... not used checking size of void *... (cached) 8 checking for aligned_alloc... yes checking for posix_memalign... yes checking for memory alignment option... SC_SIZEOF_VOID_P o-- | Results for libsc are o-- | F77: mpif77 | FFLAGS: -g -O2 | FC: mpif90 | FCFLAGS: -g -O2 | CPP: mpicc -E | CPPFLAGS: | CC: mpicc | CFLAGS: -g -O2 | CXX: mpicxx | CXXFLAGS:-g -O2 | LDFLAGS: | LIBS:-lgomp -llapack -lblas -lz -lm o-- checking that generated files are newer than configure... done configure: creating ./config.status config.status: creating Makefile config.status: creating Makefile.sc.pre config.status: creating Doxyfile config.status: creating src/pre_config.h config.status: executing src/sc_config.h commands config.status: creating src/sc_config.h - prefix SC for src/pre_config.h defines config.status: executing depfiles commands config.status: executing libtool commands configure: - SC - We did not find a recent lua containing the function lua_createtable. This is OK if the following does not matter to you: Including sc_lua.h in your code will abort the compilation. You can fix this by compiling a working lua and pointing LIBS to it. configure: - P4EST - We did not find a recent lua containing the function lua_createtable. This is OK if the following does not matter to you: Including sc_lua.h in your code will abort the compilation. You can fix this by compiling a working lua and pointing LIBS to it. And when I do a google search about lua_creatable, I find this: https://www.mail-archive.com/dealii@googlegroups.com/msg01914.html Does it seem to me that the installation dealii requires the inclusion of lua_creatable into the configuring of p4est first? And do I have to ask our machine's system administrators to install the lua before I install p4est, trilinos, dealii etc to ensure the successful installation for running in parallel? https://www.lua.org/manual/5.1/manual.html Sincerely, Chih-Che -- The deal.II project is located at http://www.dealii.org/ For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en --- You received this message because you are subscribed to the Google Groups "deal.II User Group" group. To unsubscribe from this group and stop receiving emails from it, send an email to dealii+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [deal.II] Dealii installation error
Hi Wolfgang, Hard to tell what's going wrong. What happens if you run it with one MPI > process (or without the 'mpirun -np 4')? Still, not working as you can see below. [chueh5851a@clogin2 step-xx]$ ./step-32 step-32:140628 terminated with signal 11 at PC=2aaac8a02bd2 SP=7fffb240. Backtrace: /pkg/openmpi/3.0.0/gcc/6.3.0/lib64/libmpi.so.40(PMPI_Comm_size+0x42)[0x2aaac8a02bd2] /home/chueh5851a/software/p4est-2.0/FAST/lib/libp4est-2.0.so (p4est_comm_parallel_env_get_info+0xf)[0x2aaac7aea8bf] /home/chueh5851a/software/p4est-2.0/FAST/lib/libp4est-2.0.so (p4est_new_ext+0xbd)[0x2aaac7ad9bcd] /home/chueh5851a/software/dealii/lib/libdeal_II.so.9.1.0-pre(_ZN6dealii8parallel11distributed13TriangulationILi2ELi2EE31copy_new_triangulation_to_p4estESt17integral_constantIiLi2EE+0xbe5)[0x2aaab26ff125] /home/chueh5851a/software/dealii/lib/libdeal_II.so.9.1.0-pre(_ZN6dealii8parallel11distributed13TriangulationILi2ELi2EE20create_triangulationERKSt6vectorINS_5PointILi2EdEESaIS6_EERKS4_INS_8CellDataILi2EEESaISC_EERKNS_11SubCellDataE+0x26)[0x2aaab272ef56] /home/chueh5851a/software/dealii/lib/libdeal_II.so.9.1.0-pre(_ZN6dealii13GridGenerator11hyper_shellILi2EEEvRNS_13TriangulationIXT_EXT_EEERKNS_5PointIXT_EdEEddjb+0x433)[0x2aaab1edf9c3] ./step-32(_ZN6Step3221BoussinesqFlowProblemILi2EE3runEv+0x62)[0x4c9822] ./step-32(main+0x8f)[0x474f9f] /usr/lib64/libc.so.6(__libc_start_main+0xf5)[0x2aaac9489b35] ./step-32[0x4751da] > What happens with any of the > other parallel tutorial programs -- say, step-40? > I get the same error (i.e. segmentation error). [chueh5851a@clogin2 step-xx40]$ mpirun -np 4 ./step-40 Running with Trilinos on 4 MPI rank(s)... Cycle 0: [clogin2:147265] *** Process received signal *** [clogin2:147265] Signal: Segmentation fault (11) [clogin2:147265] Signal code: Address not mapped (1) [clogin2:147265] Failing at address: 0xc8dce218 [clogin2:147264] *** Process received signal *** [clogin2:147266] *** Process received signal *** [clogin2:147266] Signal: Segmentation fault (11) [clogin2:147266] Signal code: Address not mapped (1) [clogin2:147266] Failing at address: 0xc8dce218 [clogin2:147264] Signal: Segmentation fault (11) [clogin2:147264] Signal code: Address not mapped (1) [clogin2:147264] Failing at address: 0xc8dce218 [clogin2:147263] *** Process received signal *** [clogin2:147263] Signal: Segmentation fault (11) [clogin2:147263] Signal code: Address not mapped (1) [clogin2:147263] Failing at address: 0xc8dce218 [clogin2:147265] [ 0] [clogin2:147264] [ 0] [clogin2:147266] [ 0] /usr/lib64/libc.so.6(+0x35250)[0x2aaac949d250] [clogin2:147265] [ 1] /usr/lib64/libc.so.6(+0x35250)[0x2aaac949d250] [clogin2:147266] [ 1] [clogin2:147263] [ 0] /usr/lib64/libc.so.6(+0x35250)[0x2aaac949d250] [clogin2:147264] [ 1] /pkg/openmpi/3.0.0/gcc/6.3.0/lib64/libmpi.so.40(PMPI_Comm_size+0x42)[0x2aaac8a02bd2] [clogin2:147265] [ 2] /home/chueh5851a/software/p4est-2.0/FAST/lib/ libp4est-2.0.so(p4est_comm_parallel_env_get_info+0xf)[0x2aaac7aea8bf] [clogin2:147265] [ 3] /pkg/openmpi/3.0.0/gcc/6.3.0/lib64/libmpi.so.40(PMPI_Comm_size+0x42)[0x2aaac8a02bd2] [clogin2:147266] [ 2] /home/chueh5851a/software/p4est-2.0/FAST/lib/ libp4est-2.0.so(p4est_new_ext+0xbd)[0x2aaac7ad9bcd] [clogin2:147265] [ 4] /pkg/openmpi/3.0.0/gcc/6.3.0/lib64/libmpi.so.40(PMPI_Comm_size+0x42)[0x2aaac8a02bd2] [clogin2:147264] [ 2] /home/chueh5851a/software/p4est-2.0/FAST/lib/ libp4est-2.0.so(p4est_comm_parallel_env_get_info+0xf)[0x2aaac7aea8bf] [clogin2:147264] [ 3] /home/chueh5851a/software/p4est-2.0/FAST/lib/ libp4est-2.0.so(p4est_comm_parallel_env_get_info+0xf)[0x2aaac7aea8bf] [clogin2:147266] [ 3] /home/chueh5851a/software/p4est-2.0/FAST/lib/ libp4est-2.0.so(p4est_new_ext+0xbd)[0x2aaac7ad9bcd] [clogin2:147264] [ 4] /usr/lib64/libc.so.6(+0x35250)[0x2aaac949d250] [clogin2:147263] [ 1] /home/chueh5851a/software/p4est-2.0/FAST/lib/ libp4est-2.0.so(p4est_new_ext+0xbd)[0x2aaac7ad9bcd] [clogin2:147266] [ 4] /pkg/openmpi/3.0.0/gcc/6.3.0/lib64/libmpi.so.40(PMPI_Comm_size+0x42)[0x2aaac8a02bd2] [clogin2:147263] [ 2] /home/chueh5851a/software/p4est-2.0/FAST/lib/ libp4est-2.0.so(p4est_comm_parallel_env_get_info+0xf)[0x2aaac7aea8bf] [clogin2:147263] [ 3] /home/chueh5851a/software/p4est-2.0/FAST/lib/ libp4est-2.0.so(p4est_new_ext+0xbd)[0x2aaac7ad9bcd] [clogin2:147263] [ 4] /home/chueh5851a/software/dealii/lib/libdeal_II.so.9.1.0-pre(_ZN6dealii8parallel11distributed13TriangulationILi2ELi2EE31copy_new_triangulation_to_p4estESt17integral_constantIiLi2EE+0xbe5)[0x2aaab26ff125] [clogin2:147265] [ 5] /home/chueh5851a/software/dealii/lib/libdeal_II.so.9.1.0-pre(_ZN6dealii8parallel11distributed13TriangulationILi2ELi2EE31copy_new_triangulation_to_p4estESt17integral_constantIiLi2EE+0xbe5)[0x2aaab26ff125] [clogin2:147266] [ 5] /home/chueh5851a/software/dealii/lib/libdeal_II.so.9.1.0-pre(_ZN6dealii8parallel11distributed13TriangulationILi2ELi2EE31copy_new_triangulation_to_p4estESt17integral_constantIiLi2EE+0xbe5)[0x2aaab26ff125] [clogin2:147264] [ 5]
Re: [deal.II] Dealii installation error
:87396] [ 5] /home/chueh5851a/software/dealii/lib/libdeal_II.so.9.1.0-pre(_ZN6dealii8parallel11distributed13TriangulationILi2ELi2EE31copy_new_triangulation_to_p4estESt17integral_constantIiLi2EE+0xbe5)[0x2aaab26ff125] [clogin2:87394] [ 5] /home/chueh5851a/software/dealii/lib/libdeal_II.so.9.1.0-pre(_ZN6dealii8parallel11distributed13TriangulationILi2ELi2EE31copy_new_triangulation_to_p4estESt17integral_constantIiLi2EE+0xbe5)[0x2aaab26ff125] [clogin2:87397] [ 5] /home/chueh5851a/software/dealii/lib/libdeal_II.so.9.1.0-pre(_ZN6dealii8parallel11distributed13TriangulationILi2ELi2EE31copy_new_triangulation_to_p4estESt17integral_constantIiLi2EE+0xbe5)[0x2aaab26ff125] [clogin2:87395] [ 5] /home/chueh5851a/software/dealii/lib/libdeal_II.so.9.1.0-pre(_ZN6dealii8parallel11distributed13TriangulationILi2ELi2EE20create_triangulationERKSt6vectorINS_5PointILi2EdEESaIS6_EERKS4_INS_8CellDataILi2EEESaISC_EERKNS_11SubCellDataE+0x26)[0x2aaab272ef56] [clogin2:87397] [ 6] /home/chueh5851a/software/dealii/lib/libdeal_II.so.9.1.0-pre(_ZN6dealii8parallel11distributed13TriangulationILi2ELi2EE20create_triangulationERKSt6vectorINS_5PointILi2EdEESaIS6_EERKS4_INS_8CellDataILi2EEESaISC_EERKNS_11SubCellDataE+0x26)[0x2aaab272ef56] [clogin2:87396] [ 6] /home/chueh5851a/software/dealii/lib/libdeal_II.so.9.1.0-pre(_ZN6dealii8parallel11distributed13TriangulationILi2ELi2EE20create_triangulationERKSt6vectorINS_5PointILi2EdEESaIS6_EERKS4_INS_8CellDataILi2EEESaISC_EERKNS_11SubCellDataE+0x26)[0x2aaab272ef56] [clogin2:87394] [ 6] /home/chueh5851a/software/dealii/lib/libdeal_II.so.9.1.0-pre(_ZN6dealii8parallel11distributed13TriangulationILi2ELi2EE20create_triangulationERKSt6vectorINS_5PointILi2EdEESaIS6_EERKS4_INS_8CellDataILi2EEESaISC_EERKNS_11SubCellDataE+0x26)[0x2aaab272ef56] [clogin2:87395] [ 6] /home/chueh5851a/software/dealii/lib/libdeal_II.so.9.1.0-pre(_ZN6dealii13GridGenerator11hyper_shellILi2EEEvRNS_13TriangulationIXT_EXT_EEERKNS_5PointIXT_EdEEddjb+0x433)[0x2aaab1edf9c3] [clogin2:87394] [ 7] /home/chueh5851a/software/dealii/lib/libdeal_II.so.9.1.0-pre(_ZN6dealii13GridGenerator11hyper_shellILi2EEEvRNS_13TriangulationIXT_EXT_EEERKNS_5PointIXT_EdEEddjb+0x433)[0x2aaab1edf9c3] [clogin2:87397] [ 7] /home/chueh5851a/software/dealii/lib/libdeal_II.so.9.1.0-pre(_ZN6dealii13GridGenerator11hyper_shellILi2EEEvRNS_13TriangulationIXT_EXT_EEERKNS_5PointIXT_EdEEddjb+0x433)[0x2aaab1edf9c3] [clogin2:87396] [ 7] /home/chueh5851a/software/dealii/lib/libdeal_II.so.9.1.0-pre(_ZN6dealii13GridGenerator11hyper_shellILi2EEEvRNS_13TriangulationIXT_EXT_EEERKNS_5PointIXT_EdEEddjb+0x433)[0x2aaab1edf9c3] [clogin2:87395] [ 7] ./step-32(_ZN6Step3221BoussinesqFlowProblemILi2EE3runEv+0x62)[0x4c9822] [clogin2:87395] [ 8] ./step-32(_ZN6Step3221BoussinesqFlowProblemILi2EE3runEv+0x62)[0x4c9822] [clogin2:87396] [ 8] ./step-32(_ZN6Step3221BoussinesqFlowProblemILi2EE3runEv+0x62)[0x4c9822] [clogin2:87394] [ 8] ./step-32(main+0x8f)[0x474f9f] [clogin2:87394] ./step-32(_ZN6Step3221BoussinesqFlowProblemILi2EE3runEv+0x62)[0x4c9822] [clogin2:87397] [ 8] ./step-32(main+0x8f)[0x474f9f] [clogin2:87395] [ 9] /usr/lib64/libc.so.6(__libc_start_main+0xf5)[0x2aaac9489b35] [clogin2:87395] [10] ./step-32(main+0x8f)[0x474f9f] [clogin2:87396] [ 9] /usr/lib64/libc.so.6(__libc_start_main+0xf5)[0x2aaac9489b35] [clogin2:87396] [10] [ 9] /usr/lib64/libc.so.6(__libc_start_main+0xf5)[0x2aaac9489b35] [clogin2:87394] [10] ./step-32[0x4751da] ./step-32(main+0x8f)[0x474f9f] [clogin2:87397] [ 9] /usr/lib64/libc.so.6(__libc_start_main+0xf5)[0x2aaac9489b35] [clogin2:87397] [10] [clogin2:87394] *** End of error message *** ./step-32[0x4751da] [clogin2:87395] *** End of error message *** ./step-32[0x4751da] [clogin2:87396] *** End of error message *** ./step-32[0x4751da] [clogin2:87397] *** End of error message *** --- Primary job terminated normally, but 1 process returned a non-zero exit code. Per user-direction, the job has been aborted. --- -- mpirun noticed that process rank 0 with PID 0 on node clogin2 exited on signal 11 (Segmentation fault). -- Sincerely, Chih-Che On Tue, Oct 23, 2018 at 12:16 PM Wolfgang Bangerth wrote: > On 10/22/18 8:09 AM, Chih-Che Chueh wrote: > > -- Could NOT find MPI_C (missing: MPI_C_WORKS) > > -- Could NOT find MPI_CXX (missing: MPI_CXX_WORKS) > > -- Could NOT find MPI_Fortran (missing: MPI_Fortran_WORKS) > > -- Could NOT find MPI (missing: MPI_C_FOUND MPI_CXX_FOUND > MPI_Fortran_FOUND) > > -- Performing Test MPI_HAVE_MPI_SEEK_SET > > -- Performing Test MPI_HAVE_MPI_SEEK_SET - Failed > > -- MPI_MPI_H not found! Call: > > Chih-Che, > Well, it apparently can't find anything about the MPI system. Can you take > a > look into the CMakeFiles/CMakeErrors.log file to see what the e
[deal.II] Dealii installation error
Dear all, In order to make install the deal.II, I load all the necessary modules first so that I have this [chueh5851a@clogin2 build]$ module list Currently Loaded Modulefiles: 1) blacs/openmpi/gcc/64/1.1patch03 4) scalapack/openmpi/gcc/64/2.0.2 7) mlib/OpenBLAS-0.2.20 2) lapack/gcc/64/3.7.0 5) blas/gcc/64/3.7.0 8) gcc/6.3.0 3) mpi/openmpi-3.0.0/intel2018u1 6) openmpi/gcc/64/1.10.4 9) openmpi/open64/64/1.10.3 Then I get an error after I type this. [chueh5851a@clogin2 build]$ /usr/bin/cmake3 -DCMAKE_INSTALL_PREFIX=/home/chueh5851a/software/dealii -DP4EST_DIR=/home/chueh5851a/software/p4est-2.0/FAST -DTRILINOS_DIR=/home/chueh5851a/software/Trilinos -D DEAL_II_WITH_MPI:BOOL=ON -D DEAL_II_WITH_THREADS:BOOL=ON -D DEAL_II_WITH_UMFPACK:BOOL=ON -D DEAL_II_WITH_TRILINOS:BOOL=ON -D DEAL_II_WITH_PETSC:BOOL=OFF -D DEAL_II_WITH_SLEPC:BOOL=OFF `pwd`/.. Basically, what the error is caused by not finding MPI (see the error below). But I already load the necessary modules including MPI already. Does anyone help me out? -- Include /home/chueh5851a/software/dealii/cmake/checks/check_01_cxx_features.cmake -- DEAL_II_WITH_CXX14 successfully set up -- -- Include /home/chueh5851a/software/dealii/cmake/checks/check_02_compiler_features.cmake -- -- Include /home/chueh5851a/software/dealii/cmake/checks/check_02_system_features.cmake -- -- Include /home/chueh5851a/software/dealii/cmake/checks/check_03_compiler_bugs.cmake -- -- Include /home/chueh5851a/software/dealii/cmake/configure/configure_1_cuda.cmake -- DEAL_II_WITH_CUDA is set to off. -- -- Include /home/chueh5851a/software/dealii/cmake/configure/configure_1_lapack.cmake -- A library with LAPACK API found. -- LAPACK_LIBRARIES: /usr/lib64/liblapack.so;/usr/lib64/libblas.so;/usr/lib64/libblas.so;gfortran;m;gcc_s;gcc;quadmath;m;gcc_s;gcc;c;gcc_s;gcc -- LAPACK_LINKER_FLAGS: -- Found LAPACK -- DEAL_II_WITH_LAPACK successfully set up with external dependencies. -- -- Include /home/chueh5851a/software/dealii/cmake/configure/configure_1_mpi.cmake -- Could NOT find MPI_C (missing: MPI_C_WORKS) -- Could NOT find MPI_CXX (missing: MPI_CXX_WORKS) -- Could NOT find MPI_Fortran (missing: MPI_Fortran_WORKS) -- Could NOT find MPI (missing: MPI_C_FOUND MPI_CXX_FOUND MPI_Fortran_FOUND) -- Performing Test MPI_HAVE_MPI_SEEK_SET -- Performing Test MPI_HAVE_MPI_SEEK_SET - Failed -- MPI_MPI_H not found! Call: -- FIND_FILE(MPI_MPI_H NAMES mpi.h HINTS) -- MPI_VERSION: 0.0 -- MPI_LIBRARIES: /pkg/openmpi/3.0.0/intel/2018_u1/lib64/libmpi.so;/pkg/openmpi/3.0.0/intel/2018_u1/lib64/libmpi_usempif08.so;/pkg/openmpi/3.0.0/intel/2018_u1/lib64/libmpi_usempi_ignore_tkr.so;/pkg/openmpi/3.0.0/intel/2018_u1/lib64/libmpi_mpifh.so;/pkg/openmpi/3.0.0/intel/2018_u1/lib64/libmpi.so;/pkg/openmpi/3.0.0/intel/2018_u1/lib64/libmpi.so -- MPI_INCLUDE_DIRS: -- MPI_USER_INCLUDE_DIRS: -- MPI_CXX_FLAGS: -- MPI_LINKER_FLAGS: -Wl,-rpath -Wl,/opt/pbs/lib -Wl,-rpath -Wl,/usr/lib64 -Wl,-rpath -Wl,/pkg/openmpi/3.0.0/intel/2018_u1/lib64 -Wl,--enable-new-dtags -L/opt/pbs/lib -L/usr/lib64 -pthread -- Could NOT find MPI -- DEAL_II_WITH_MPI has unmet external dependencies. CMake Error at cmake/configure/configure_1_mpi.cmake:84 (MESSAGE): Could not find any suitable mpi library! Please ensure that an mpi library is installed on your computer and set CMAKE_CXX_COMPILER to the appropriate mpi wrappers: $ CXX=".../mpicxx" cmake <...> $ cmake -DCMAKE_CXX_COMPILER=".../mpicxx" <...> Or with additional C and Fortran wrappers (recommended!): $ CC=".../mpicc" CXX=".../mpicxx" F90=".../mpif90" cmake <...> $ cmake -DCMAKE_C_COMPILER=".../mpicc"\ -DCMAKE_CXX_COMPILER=".../mpicxx"\ -DCMAKE_Fortran_COMPILER=".../mpif90"\ <...> Call Stack (most recent call first): build/CMakeFiles/CMakeTmp/evaluate_expression.tmp:1 (FEATURE_MPI_ERROR_MESSAGE) cmake/macros/macro_evaluate_expression.cmake:30 (INCLUDE) cmake/macros/macro_configure_feature.cmake:267 (EVALUATE_EXPRESSION) cmake/configure/configure_1_mpi.cmake:101 (CONFIGURE_FEATURE) cmake/macros/macro_verbose_include.cmake:19 (INCLUDE) CMakeLists.txt:124 (VERBOSE_INCLUDE) -- Configuring incomplete, errors occurred! See also "/home/chueh5851a/software/dealii/build/CMakeFiles/CMakeOutput.log". See also "/home/chueh5851a/software/dealii/build/CMakeFiles/CMakeError.log". -- The deal.II project is located at http://www.dealii.org/ For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en --- You received this message because you are subscribed to the Google Groups "deal.II User Group" group. To unsubscribe from this group and stop receiving emails from it, send an email to dealii+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [deal.II] Re: about deal.II with CUDA C programming acceleration
> I'm going to add that installing a compiler will really only be the >> first >> step. After that, of course, comes the task of understanding how to >> program these graphics cards, and all of the other things that require >> interacting with the appropriate software levels. You will need to >> master >> all of these things at one point or other. You might as well learn >> how to >> install operating systems and compilers. >> >> >> >> Frankly speaking, all we need is write a simple, clear instruction on the >> updating the newer version of GNU C like you in deal.II write this for >> p4est (http://www.dealii.org/developer/external-libs/p4est.html) so that >> deal.II users would follow easily and quickly. Is it possible for any one >> in deal.II to have this kind of information? >> > That, too, is asking for too much. It's not something the average deal.II > user ever needs to do, and it's something that we will never be able to > document as well as the GCC folks. Please read their instructions -- they > have very good installation instructions. > OK! I see. I will solve the problem I face now. Thanks for the answer! Chih-Che -- The deal.II project is located at http://www.dealii.org/ For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en --- You received this message because you are subscribed to the Google Groups "deal.II User Group" group. To unsubscribe from this group and stop receiving emails from it, send an email to dealii+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [deal.II] Re: about deal.II with CUDA C programming acceleration
Hi Wolfgang, I think you are expecting too much of Bruno. He has a day job as well, for > which he is paid. I think you will need to figure out the installation on > your system yourself -- otherwise, Bruno (or any other among us) will be > asked to do so for everyone else on the mailing list who wants to install > CUDA. > OK! I understand this now. Hi Bruno, Please ignore this request. Sorry about that. > I'm going to add that installing a compiler will really only be the first > step. After that, of course, comes the task of understanding how to program > these graphics cards, and all of the other things that require interacting > with the appropriate software levels. You will need to master all of these > things at one point or other. You might as well learn how to install > operating systems and compilers. > Frankly speaking, all we need is write a simple, clear instruction on the updating the newer version of GNU C like you in deal.II write this for p4est (http://www.dealii.org/developer/external-libs/p4est.html) so that deal.II users would follow easily and quickly. Is it possible for any one in deal.II to have this kind of information? Sincerely, Chih-Che -- The deal.II project is located at http://www.dealii.org/ For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en --- You received this message because you are subscribed to the Google Groups "deal.II User Group" group. To unsubscribe from this group and stop receiving emails from it, send an email to dealii+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [deal.II] Re: about deal.II with CUDA C programming acceleration
Hi Bruno, > You definitely don't want to update your OS. I am also using ubuntu and I > have several versions of gcc and clang installed. Lately, I have been using > spack to install everything for me (https://github.com/LLNL/spack). This > is becoming the standard way to install new programs on clusters at the DOE > labs. Last time I installed gcc myself was about a year ago and I just > followed the instruction here https://gcc.gnu.org/install/configure.html > If your admin wants to shoot me an email with the errors he is getting, I > will be happy to help. > Thanks for your kind help. I have conveyed what you say above to our system administrator, who will surely follow the instruction you provide. If he has any trouble with this and need your help, he will tell me (as his English is not as good as mine) and then I will let you know. Sincerely, Chih-Che -- The deal.II project is located at http://www.dealii.org/ For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en --- You received this message because you are subscribed to the Google Groups "deal.II User Group" group. To unsubscribe from this group and stop receiving emails from it, send an email to dealii+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [deal.II] Re: about deal.II with CUDA C programming acceleration
HI Bruno, Thanks for the reminder and helps. Indeed, our system administrator has trouble with installing the newer version of GNU C compiler. Would you provide detailed instructions on how to installing the GNU C compiler? Thanks! The system administrator told me that he was trying to install the newer version of GNU C compiler alone, but failed. He told me that people usually reinstall the newer version of linux UBUNTU system with the newer version of GNU C compiler instead of installing the newer version of GNU C compiler alone. However, because we have other important softwares installed/used in our system, I don't know when he is going to update the entire linux system. So, I guess that the only way to get the newer version of GNU C compiler is feasible right now. But we don't know how to get it working. Sincerely, Chih-Che On Sat, Sep 16, 2017 at 2:15 AM, Bruno Turcksin <bruno.turck...@gmail.com> wrote: > Chih-Che, > > Did you get everything to work or you gave up? You are the first > non-developer trying to install deal.II with CUDA, so I am really > interested to know which problems you are encountering. > > Best, > > Bruno > > > 2017-08-23 20:49 GMT-04:00 Bruno Turcksin <bruno.turck...@gmail.com>: > >> Chih-Che, >> >> 2017-08-23 20:09 GMT-04:00 Chih-Che Chueh <chue...@gmail.com>: >> > That's still not working with getting the same error. Is this just >> because >> > it can't find the BLAS and LAPACK libraies? >> Did you reconfigure Trilinos in an empty directory? If not do it, you >> never know what has been cached by CMake... >> >> The problem with MKL is that instead of having one or two libraries, >> they have split BLAS and LAPACK into a bunch of small libraries and >> you need to find which libraries you need to include. You can use this >> website to help you: >> https://software.intel.com/en-us/articles/intel-mkl-link-line-advisor >> Look at the libraries that are needed at linking time and add them to >> BLAS_LIBRARY_NAMES or LAPACK_LIBRARY_NAMES. It shouldn't matter which >> one. >> >> > By the way, before I installed the trilinos, I module load many things: >> > >> > [chueh@icy]$ module load intel/comp-15.0.0 >> > [chueh@icy]$ module load intel/mkl-11.1.2 >> > [chueh@icy]$ module load openmpi-1.6.5/intel-15.0 >> Maybe pgi was a better idea :-D There are two problems with intel 15: >> 1) we don't test deal.II with intel15 so I am not sure if it will >> works but I guess it should work >> 2) the big problem is how are you going to use CUDA? I didn't know >> that nvcc supported icc but apparently it kinda does but it looks >> pretty bad https://devtalk.nvidia.com/default/topic/947888/intel-icc- >> compiler-and-c-11/?offset=4 >> For your information, deal now requires C++11 so it won't work. Don't >> you have gcc 5 on your machine? >> >> > [chueh@icy ~]$ module av >> > >> > >> --- >> > /work2/opt/Modules/modulefiles >> > >> --- >> > cmake/3.9.1hdf5/gcc-4.4.6_1.8.14 >> > intel/comp-15.0module-git >> > ncl_ncarg/6.3.0netcdf/hdf5_pgi-15.1_4.3.3.1 >> > pgi/15.10/x86_64 >> > dothdf5/intel-15.0_1.10.0 >> > intel/comp-15.0.0 module-info >> > netcdf/hdf5_gnu_4.3.3.1null >> pgi_mpi/15.10 >> > grads/2.0.2hdf5/intel-15.0_1.8.14 >> intel/mkl-11.1 >> > modulesnetcdf/hdf5_intel-15.0_4.3.3.1 >> > opengrads/2.0.2.oga.2 python/3.5.3 >> > grads/2.1.a3 hdf5/pgi-15.1_1.8.14 >> > intel/mkl-11.1.2 ncl_ncarg/6.1.2 >> > netcdf/hdf5_intel-15.0_4.4.1.1 openmpi-1.6.5/intel-15.0 use.own >> Are these all the modules that are available to you :-( Ideally, you >> have something like openmpi/gcc5 (gcc 4.8 or 4.9 would also work not 6 >> or later because they are not compatible with CUDA8). Now your problem >> is that pgi-15 is probably too old to be useful for C++ code and >> intel-15 does not work well with cuda. Personally, I would compile my >> own gcc 5.4. It's actually pretty easy because you only need C++ and >> C. You should _not_ compile support for the other languages. Then, you >> load openmpi-1.6.5/intel-15.0 and use this >> https://www.open-mpi.org/faq/?category=mpi-apps#ov
Re: [deal.II] Re: about deal.II with CUDA C programming acceleration
Hi Bruno > I have already moved to the trilinos installation. But after installing > for > > a while, I get the error that you can see below. Is this because I am > using > > an old version of Intel MKL Math Kernel library? If not, does any one > know > > how to fix it? > First you used pgi now you want to use MKL, I see that you like challenges > ;-) > Believe me! I don't really like this kind of challenge. I really just want to fix the problem and get to work on CUDA programming. ;-) > > -DBLAS_LIBRARY_NAMES:STRING="mkl_intel_lp64" \ > > > > -DBLAS_LIBRARY_DIRS:STRING=/work2/opt/intel/composer_xe_2013 > _sp1.2.144/mkl/lib/intel64 > > \ > > -DLAPACK_LIBRARY_NAMES:STRING="mkl_intel_lp64" \ > > > > -DLAPACK_LIBRARY_DIRS:STRING=/work2/opt/intel/composer_xe_20 > 13_sp1.2.144/mkl/lib/intel64 > > \ > Try using this for the library names > -DBLAS_LIBRARY_NAMES:STRING="mkl_core;mkl_sequential" > -DLAPACK_LIBRARY_NAMES:STRING="mkl_intel_lp64" > That's still not working with getting the same error. Is this just because it can't find the BLAS and LAPACK libraies? By the way, before I installed the trilinos, I module load many things: [chueh@icy]$ module load intel/comp-15.0.0 [chueh@icy]$ module load intel/mkl-11.1.2 [chueh@icy]$ module load openmpi-1.6.5/intel-15.0 [chueh@icy]$ module load cmake/3.9.1 Can you help me see if there is any other things I have to module load before the trilinos installation to get to work successfully? Thanks! [chueh@icy ~]$ module av --- /work2/opt/Modules/modulefiles -- - cmake/3.9.1hdf5/gcc-4.4.6_1.8.14 intel/comp-15.0module-git ncl_ncarg/6.3.0netcdf/hdf5_pgi-15.1_4.3.3.1 pgi/15.10/x86_64 dothdf5/intel-15.0_1.10.0 intel/comp-15.0.0 module-info netcdf/hdf5_gnu_4.3.3.1null pgi_mpi/15.10 grads/2.0.2hdf5/intel-15.0_1.8.14 intel/mkl-11.1 modules netcdf/hdf5_intel-15.0_4.3.3.1 opengrads/2.0.2.oga.2 python/3.5.3 grads/2.1.a3 hdf5/pgi-15.1_1.8.14 intel/mkl-11.1.2 ncl_ncarg/6.1.2 netcdf/hdf5_intel-15.0_4.4.1.1 openmpi-1.6.5/intel-15.0 use.own Best, Chih-Che -- The deal.II project is located at http://www.dealii.org/ For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en --- You received this message because you are subscribed to the Google Groups "deal.II User Group" group. To unsubscribe from this group and stop receiving emails from it, send an email to dealii+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [deal.II] Re: about deal.II with CUDA C programming acceleration
Hi Daniel, Thanks for the answer for the successful p4est installation. I have not seen you since 2013 deal.II workshop. It was nice to meet you and eat dinner with you with the same table. Hi Daniel, Bruno or any others, I have already moved to the trilinos installation. But after installing for a while, I get the error that you can see below. Is this because I am using an old version of Intel MKL Math Kernel library? If not, does any one know how to fix it? cmake\ -DTrilinos_ENABLE_Amesos=ON \ -DTrilinos_ENABLE_Epetra=ON \ -DTrilinos_ENABLE_Ifpack=ON \ -DTrilinos_ENABLE_AztecOO=ON \ -DTrilinos_ENABLE_Sacado=ON \ -DTrilinos_ENABLE_Teuchos=ON \ -DTrilinos_ENABLE_MueLu=ON \ -DTrilinos_ENABLE_ML=ON \ -DTrilinos_ENABLE_EXPLICIT_INSTANTIATION=ON \ -DTrilinos_VERBOSE_CONFIGURE=OFF \ -DBLAS_LIBRARY_NAMES:STRING="mkl_intel_lp64" \ -DBLAS_LIBRARY_DIRS:STRING=/work2/opt/intel/composer_xe_2013_sp1.2.144/ mkl/lib/intel64 \ -DLAPACK_LIBRARY_NAMES:STRING="mkl_intel_lp64" \ -DLAPACK_LIBRARY_DIRS:STRING=/work2/opt/intel/composer_xe _2013_sp1.2.144/mkl/lib/intel64 \ -DTPL_ENABLE_MPI=ON \ -DBUILD_SHARED_LIBS=ON \ -DCMAKE_VERBOSE_MAKEFILE=OFF \ -DCMAKE_BUILD_TYPE=RELEASE \ -DCMAKE_INSTALL_PREFIX:PATH=/work2/chueh/deal.II_project/trilinos \ ../ make install ==ERROR [...] /work2/opt/intel/composer_xe_2013_sp1.2.144/mkl/lib/intel64/libmkl_intel_lp64.so: undefined reference to `mkl_lapack_dlanv2' /work2/opt/intel/composer_xe_2013_sp1.2.144/mkl/lib/intel64/libmkl_intel_lp64.so: undefined reference to `mkl_lapack_sla_gercond' /work2/opt/intel/composer_xe_2013_sp1.2.144/mkl/lib/intel64/libmkl_intel_lp64.so: undefined reference to `mkl_trs_dtrnlsp_init' /work2/opt/intel/composer_xe_2013_sp1.2.144/mkl/lib/intel64/libmkl_intel_lp64.so: undefined reference to `mkl_spblas_lp64_mkl_sbsrsv' /work2/opt/intel/composer_xe_2013_sp1.2.144/mkl/lib/intel64/libmkl_intel_lp64.so: undefined reference to `mkl_blas_dsyr2k' /work2/opt/intel/composer_xe_2013_sp1.2.144/mkl/lib/intel64/libmkl_intel_lp64.so: undefined reference to `mkl_lapack_zlantp' /work2/opt/intel/composer_xe_2013_sp1.2.144/mkl/lib/intel64/libmkl_intel_lp64.so: undefined reference to `mkl_spblas_lp64_mkl_zdiamm' /work2/opt/intel/composer_xe_2013_sp1.2.144/mkl/lib/intel64/libmkl_intel_lp64.so: undefined reference to `mkl_lapack_dgelsd' make[2]: *** [packages/amesos2/src/SuiteSparse /KLU2/Source/Amesos2_klu2_simple.exe] Error 1 make[1]: *** [packages/amesos2/src/SuiteSparse/KLU2/Source/CMakeFiles /Amesos2_klu2_simple.dir/all] Error 2 make: *** [all] Error 2 Sincerely, Chih-Che On Wed, Aug 23, 2017 at 5:21 PM, Daniel Arndtwrote: > Chih-Che, > > [...] >> Build FAST version in /work2/chueh/deal.II_project/p4est/p4est-build/FAST >> configure: WARNING: Static source code checker splint not found >> configure: WARNING: Static source code checker splint not found >> FAST version installed in /work2/chueh/deal.II_project/p4est/FAST >> >> Build DEBUG version in /work2/chueh/deal.II_project/p >> 4est/p4est-build/DEBUG >> configure: WARNING: Static source code checker splint not found >> configure: WARNING: Static source code checker splint not found >> DEBUG version installed in /work2/chueh/deal.II_project/p4est/DEBUG >> >> Can you help me tell me if p4est has been installed successfully? >> > Yes, this looks good. The next step is to tell deal.II via > cmake -DP4EST_DIR=/work2/chueh/deal.II_project/p4est/ . > to use P4EST and check if it works. > > Best, > Daniel > > -- > The deal.II project is located at http://www.dealii.org/ > For mailing list/forum options, see https://groups.google.com/d/ > forum/dealii?hl=en > --- > You received this message because you are subscribed to the Google Groups > "deal.II User Group" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to dealii+unsubscr...@googlegroups.com. > For more options, visit https://groups.google.com/d/optout. > -- The deal.II project is located at http://www.dealii.org/ For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en --- You received this message because you are subscribed to the Google Groups "deal.II User Group" group. To unsubscribe from this group and stop receiving emails from it, send an email to dealii+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [deal.II] Re: trilino installation problem
Hi Jean-Paul, You are right. It is not matching what I expected. I am going to ask our system administrator to fix it. Thanks! [chueh@icy build]$ cmake --version cmake version 2.6-patch 4 [chueh@icy build]$ which cmake /usr/bin/cmake Sincerely, Chih-Che On Wed, Aug 23, 2017 at 3:38 PM, Jean-Paul Pelteret <jppelte...@gmail.com> wrote: > Hi Chih-Che, > > I suppose the obvious first question is what the output of the following > two commands are > > which cmake > cmake --version > > and whether or not those match that expected after you've load the CMake > module. > > Regards, > Jean-Paul > > > On Wednesday, August 23, 2017 at 9:20:41 AM UTC+2, Chih-Che Chueh wrote: >> >> Hi all, >> >> When I try to install trilinos with Cmake, I find something strange: >> while I am using cmake/3.9.1, after the installation, we get the error. But >> in reality the version I am using is newer than 2.6.4. Does any one who >> faced the same problem know how to fix the problem? >> >> >> >> [chueh@icy build]$ module load cmake/3.9.1 >> [chueh@icy build]$ cmake\ >> > -DTrilinos_ENABLE_Amesos=ON \ >> > -DTrilinos_ENABLE_Epetra=ON \ >> > -DTrilinos_ENABLE_Ifpack=ON \ >> > -DTrilinos_ENABLE_AztecOO=ON \ >> > -DTrilinos_ENABLE_Sacado=ON \ >> > -DTrilinos_ENABLE_Teuchos=ON \ >> > -DTrilinos_ENABLE_MueLu=ON \ >> > -DTrilinos_ENABLE_ML=ON \ >> > -DTrilinos_VERBOSE_CONFIGURE=OFF \ >> > -DTPL_ENABLE_MPI=ON \ >> > -DBUILD_SHARED_LIBS=ON \ >> > -DCMAKE_VERBOSE_MAKEFILE=OFF \ >> > -DCMAKE_BUILD_TYPE=RELEASE \ >> > -DCMAKE_INSTALL_PREFIX:PATH=/work2/chueh/deal.II_project/trilinos \ >> > ../ >> CMake Error at CMakeLists.txt:62 (CMAKE_MINIMUM_REQUIRED): >> CMake 2.8.11 or higher is required. You are running version 2.6.4 >> >> >> -- Configuring incomplete, errors occurred! >> >> Best, >> >> Chih-Che >> >> -- > The deal.II project is located at http://www.dealii.org/ > For mailing list/forum options, see https://groups.google.com/d/ > forum/dealii?hl=en > --- > You received this message because you are subscribed to the Google Groups > "deal.II User Group" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to dealii+unsubscr...@googlegroups.com. > For more options, visit https://groups.google.com/d/optout. > -- The deal.II project is located at http://www.dealii.org/ For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en --- You received this message because you are subscribed to the Google Groups "deal.II User Group" group. To unsubscribe from this group and stop receiving emails from it, send an email to dealii+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [deal.II] Re: about deal.II with CUDA C programming acceleration
Bruno, > pgc++-Error-Unknown switch: -print-prog-name=ld > Don't use pgi :-) We aren't done with the pgi port. > I just turn off pgi on my linux background while turning on the necessary intel C compiler etc, and try again to use the online script. It seems to work well but with some warnings. [chueh@icy p4est]$ module load intel/comp-15.0.0 [chueh@icy p4est]$ module load intel/mkl-11.1.2 [chueh@icy p4est]$ module load openmpi-1.6.5/intel-15.0 [chueh@icy p4est]$ sh ./p4est-setup.sh p4est-1.1.tar.gz /work2/chueh/deal.II_project/p4est CFLAGS_FAST: -O2 CFLAGS_DEBUG: -O0 -g This script tries to unpack, configure and build the p4est library. Build FAST: /work2/chueh/deal.II_project/p4est/p4est-build/FAST Build DEBUG: /work2/chueh/deal.II_project/p4est/p4est-build/DEBUG Install FAST: /work2/chueh/deal.II_project/p4est/FAST Install DEBUG: /work2/chueh/deal.II_project/p4est/DEBUG Checking environment: CFLAGS P4EST_CFLAGS_FAST P4EST_CFLAGS_DEBUG Unpack directory: /work2/chueh/deal.II_project/p4est/p4est-1.1 Source directory found (remove it to unpack anew) See output in files .../config.output and .../make.output Build FAST version in /work2/chueh/deal.II_project/p4est/p4est-build/FAST configure: WARNING: Static source code checker splint not found configure: WARNING: Static source code checker splint not found FAST version installed in /work2/chueh/deal.II_project/p4est/FAST Build DEBUG version in /work2/chueh/deal.II_project/p4est/p4est-build/DEBUG configure: WARNING: Static source code checker splint not found configure: WARNING: Static source code checker splint not found DEBUG version installed in /work2/chueh/deal.II_project/p4est/DEBUG Can you help me tell me if p4est has been installed successfully? Sincerely, Chih-Che -- The deal.II project is located at http://www.dealii.org/ For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en --- You received this message because you are subscribed to the Google Groups "deal.II User Group" group. To unsubscribe from this group and stop receiving emails from it, send an email to dealii+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [deal.II] Re: about deal.II with CUDA C programming acceleration
Hi Bruno, Thanks for the notification. I am going to install p4est, trilino and the deal.II with CUDA, as I also want to run some tutorial programs made before. I will let you know if I have any problem or trouble installing them. By the way, the CUDA compiler version in our server has been updated into CUDA 8.0, as you suggested last week. [chueh@icycn06 ~]$ nvcc --version nvcc: NVIDIA (R) Cuda compiler driver Copyright (c) 2005-2016 NVIDIA Corporation Built on Tue_Jan_10_13:22:03_CST_2017 Cuda compilation tools, release 8.0, V8.0.61 Sincerely, Chih-Che On Mon, Aug 21, 2017 at 8:16 PM, Bruno Turcksin <bruno.turck...@gmail.com> wrote: > Chih-Che > > 2017-08-17 20:39 GMT-04:00 Chih-Che Chueh <chue...@gmail.com>: > > I can wait for the tests that are done. No worry. > Everything is working now. > > Best, > > Bruno > > -- > The deal.II project is located at http://www.dealii.org/ > For mailing list/forum options, see https://groups.google.com/d/fo > rum/dealii?hl=en > --- > You received this message because you are subscribed to the Google Groups > "deal.II User Group" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to dealii+unsubscr...@googlegroups.com. > For more options, visit https://groups.google.com/d/optout. > -- The deal.II project is located at http://www.dealii.org/ For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en --- You received this message because you are subscribed to the Google Groups "deal.II User Group" group. To unsubscribe from this group and stop receiving emails from it, send an email to dealii+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [deal.II] Re: about deal.II with CUDA C programming acceleration
Bruno, > Let me know once your merger is done so that I can start to install the > > deal.II with CUDA in the server I use now. > The merge was done but in the mean time we broke three tests for CUDA > :-(. I think that it is still worth to try to install deal with CUDA > and just update later. > I can wait for the tests that are done. No worry. > > The CUDA compiler I use now in the server is shown in the following: > > [chueh@icycn06 cuda]$ nvcc --version > > nvcc: NVIDIA (R) Cuda compiler driver > > Copyright (c) 2005-2015 NVIDIA Corporation > > Built on Tue_Aug_11_14:27:32_CDT_2015 > > Cuda compilation tools, release 7.5, V7.5.17 > I don't know if that will work. I think we require CUDA 8.0. Because > the feature is brand new and we are all using CUDA 8.0, I am not sure > that it will work with CUDA 7.5 (nobody has ever tried). > OK! I am going to ask our system administration to have CUDA 8 compiler version installed in the GPU before I try to install the deal.II with CUDA. > > Also, I would appreciate to see the detailed instruction on how to > install > > deal.II together with CUDA, because I don't know how to do that. > Yes, I still need to write the documentation. However it should be > pretty simple. You need CMake 3.9 and then just use > -DDEAL_II_WITH_CUDA=ON and it should work. Let me know if you have any > problem. OK! I see. Sincerely, Chih-Che -- The deal.II project is located at http://www.dealii.org/ For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en --- You received this message because you are subscribed to the Google Groups "deal.II User Group" group. To unsubscribe from this group and stop receiving emails from it, send an email to dealii+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [deal.II] Re: about deal.II with CUDA C programming acceleration
Bruno, > Is there any reason why you still used the previous way? Is this because > > your GPU accelerator was launched some years ago so that you are unable > to > > use the new simple way? > The problem with UVM is that it makes it easy to write bad code. I am > working on a code where we use UVM and it happens quite often that we > run some code that was intended for the GPU on the CPU. Because of UVM > it works but at the cost of performance and so it is harder to detect > that you made a mistake. Another advantage of copying the data > manually is that you don't hide the costly copy operation. This is > also the philosophy used in Kokkos. So yes UVM makes writing the code > easier but you need to be a lot more careful when you write the code > if you don't want to perform a bunch of useless copy. Of course, if > you have NVLink it doesn't matter but that only works for POWER chips > which most people don't have access to. > That is a convincing reason for you to do so. I know it now. Thanks! Let me know once your merger is done so that I can start to install the deal.II with CUDA in the server I use now. The CUDA compiler I use now in the server is shown in the following: [chueh@icycn06 cuda]$ nvcc --version nvcc: NVIDIA (R) Cuda compiler driver Copyright (c) 2005-2015 NVIDIA Corporation Built on Tue_Aug_11_14:27:32_CDT_2015 Cuda compilation tools, release 7.5, V7.5.17 Also, I would appreciate to see the detailed instruction on how to install deal.II together with CUDA, because I don't know how to do that. Thanks! Sincerely, Chih-Che -- The deal.II project is located at http://www.dealii.org/ For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en --- You received this message because you are subscribed to the Google Groups "deal.II User Group" group. To unsubscribe from this group and stop receiving emails from it, send an email to dealii+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [deal.II] Re: about deal.II with CUDA C programming acceleration
Bruno, It is nice to meet you again. I still remember we met and chatted a little bit in 2013 deal.II workshop, which was held in Texas A University. Thank you for letting me be involved in this development. Actually, once I really get started after the installation of the development version of deal.II in the server I am using, I am sure that I will have many questions to you about CUDA site of deal.II, and I will need your helps. After I just take a glimpse at this to try to understand the CUDA style you tried to implement, I have a quick question to you: When you did test the CUDA code, you used cudaMalloc to create an object on the device (GPU) and then copy the results from the device to the host using cudaMemcpy. As far as I know, now CUDA offers a simple way using Unified Memory, which is providing a single memory space accessible by all GPUs and CPUs in your system with the efficient page migration engine of the recently released NVIDIA Tesla P100. This implementation is like the following CUDA code. =Unified Memory== int N = 1; float *x, *y; // Allocate Unified Memory – accessible from CPU or GPU cudaMallocManaged(, N*sizeof(float)); cudaMallocManaged(, N*sizeof(float)); // initialize x and y arrays on the host for (int i = 0; i < N; i++) { x[i] = 1.0f; y[i] = 2.0f; } add<<<1, 1>>>(N, x, y); // Free memory cudaFree(x); cudaFree(y); === You might also want to take a look at this about unified memory: https://devblogs.nvidia.com/parallelforall/cuda-8-features-revealed/ Is there any reason why you still used the previous way? Is this because your GPU accelerator was launched some years ago so that you are unable to use the new simple way? Thanks! Sincerely, Chih-Che On Tue, Aug 15, 2017 at 7:19 PM, Bruno Turcksin <bruno.turck...@gmail.com> wrote: > Chih-Che, > > The CUDA support in deal.II is very new. It is only in the development > version of deal (https://github.com/dealii/dealii). You can see our > current development plan here https://github.com/dealii/dealii/projects/2 > Right now, we have support for vector and partial support for matrix-free > (a good place to see the capabilities is the test suite > https://github.com/dealii/dealii/tree/master/tests/cuda). I would advise > you to wait for this PR https://github.com/dealii/dealii/pull/4846 to be > merged before your try to install deal with CUDA. This PR makes it a lot > easier to install deal with CUDA. If you want to work on CUDA, you should > work on something that you like / is of interest to you. If you want to > help our existing effort, you can pick something from this list > https://github.com/dealii/dealii/issues/4399. I am working on the first > item but let me know if you find anything else interesting, I can help you > implementing it. > > If you have any questions, please ask. We are looking for people to help > us with CUDA. > > Best, > > Bruno > > > On Monday, August 14, 2017 at 7:56:45 AM UTC-4, Chih-Che Chueh wrote: >> >> Dear deal.II developers and users, >> >> Recently, I spent some spare time assimilating CUDA C programming in the >> last few months, and I already know very well how to use CUDA stream >> events to let CPU and kernel (GPU) execution work asynchronously with >> efficiently overlapping data transfer between CPU and GPU, how to use >> shared memory to ensure global memory coalescing efficiently, how to map >> threads to matrix elements either using CARTESIAN x, y, z or a row/column >> mapping in GPU, as well as how to use shared memory to enhance data >> reuse. Most importantly, for actual practice, we have a GPU accelerator >> (i.e. NVIDIA Tesla K40) that was bought last year. I plan to use the CUDA C >> programming to deal with big data or image identification with artificial >> intelligence (deep learning) for atmospheric data. >> >> Anyway, I am writing to ask if I could get involved with a deal.II >> project of people who are working on asynchronous adaptive mesh refinement >> for acceleration or other performance improvement in deal.II with CUDA C >> programming. >> >> Thanks! >> >> Sincerely, >> >> Chih-Che >> > -- > The deal.II project is located at http://www.dealii.org/ > For mailing list/forum options, see https://groups.google.com/d/ > forum/dealii?hl=en > --- > You received this message because you are subscribed to the Google Groups > "deal.II User Group" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to dealii+unsubscr...@googlegroups.com. > For more options, visit https://groups.google.com/d/optout. > -- T
[deal.II] about deal.II with CUDA C programming acceleration
Dear deal.II developers and users, Recently, I spent some spare time assimilating CUDA C programming in the last few months, and I already know very well how to use CUDA stream events to let CPU and kernel (GPU) execution work asynchronously with efficiently overlapping data transfer between CPU and GPU, how to use shared memory to ensure global memory coalescing efficiently, how to map threads to matrix elements either using CARTESIAN x, y, z or a row/column mapping in GPU, as well as how to use shared memory to enhance data reuse. Most importantly, for actual practice, we have a GPU accelerator (i.e. NVIDIA Tesla K40) that was bought last year. I plan to use the CUDA C programming to deal with big data or image identification with artificial intelligence (deep learning) for atmospheric data. Anyway, I am writing to ask if I could get involved with a deal.II project of people who are working on asynchronous adaptive mesh refinement for acceleration or other performance improvement in deal.II with CUDA C programming. Thanks! Sincerely, Chih-Che -- The deal.II project is located at http://www.dealii.org/ For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en --- You received this message because you are subscribed to the Google Groups "deal.II User Group" group. To unsubscribe from this group and stop receiving emails from it, send an email to dealii+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.