Open mjribeiro opened 1 year ago
@mjribeiro : I locally tried to build with NVHPC 22.11 and CUDA 11.8 but don't see obvious issue. In your logs I see:
nvc++-Fatal-/opt/nvidia/hpc_sdk/Linux_x86_64/22.11/compilers/bin/tools/cpp2 TERMINATED by signal 11
Arguments to /opt/nvidia/hpc_sdk/Linux_x86_64/22.11/compilers/bin/tools/cpp2
/opt/nvidia/hpc_sdk/Linux_x86_64/22.11/compilers/bin/tools/cpp2 /home/mafalda/nrn/src/ivoc/ivocvect.cpp ...
i.e. cpp2
seems to be failing. First thing comes to my mind is if you could try with --parallel 2
or --parallel 1
in case this is memory usage related issue from the compiler itself.
My local cmake / build log is:
$ cmake .. -DNRN_ENABLE_CORENEURON=ON -DCORENRN_ENABLE_GPU=ON -DNRN_ENABLE_INTERVIEWS=OFF -DNRN_ENABLE_RX3D=OFF -DCMAKE_INSTALL_PREFIX=$HOME/install -DCMAKE_C_COMPILER=nvc -DCMAKE_CXX_COMPILER=nvc++ -DCMAKE_CUDA_COMPILER=nvcc
-- The C compiler identification is NVHPC 22.11.0
-- The CXX compiler identification is NVHPC 22.11.0
-- Detecting C compiler ABI info
-- Detecting C compiler ABI info - done
-- Check for working C compiler: /gpfs/bbp.cscs.ch/ssd/apps/bsd/pulls/1654/stage_externals/install_gcc-11.2.0-skylake/nvhpc-22.11-rx7dzj/Linux_x86_64/22.11/compilers/bin/nvc - skipped
-- Detecting C compile features
-- Detecting C compile features - done
-- Detecting CXX compiler ABI info
-- Detecting CXX compiler ABI info - done
-- Check for working CXX compiler: /gpfs/bbp.cscs.ch/ssd/apps/bsd/pulls/1654/stage_externals/install_gcc-11.2.0-skylake/nvhpc-22.11-rx7dzj/Linux_x86_64/22.11/compilers/bin/nvc++ - skipped
-- Detecting CXX compile features
-- Detecting CXX compile features - done
-- Setting build type to 'RelWithDebInfo' as none was specified.
-- 3rd party project: using Random123 from "external/Random123"
-- -DPYTHON_EXECUTABLE not specified. Looking for `python3` in the PATH exclusively...
-- Setting PYTHON_EXECUTABLE=/gpfs/bbp.cscs.ch/ssd/apps/bsd/2022-01-10/stage_externals/install_gcc-11.2.0-skylake/python-3.9.7-yj5alh/bin/python3
-- Found BISON: /gpfs/bbp.cscs.ch/ssd/apps/bsd/2022-01-10/stage_externals/install_gcc-11.2.0-skylake/bison-3.8.2-ej2cew/bin/bison (found version "3.8.2")
-- Found FLEX: /gpfs/bbp.cscs.ch/ssd/apps/bsd/2022-01-10/stage_externals/install_gcc-11.2.0-skylake/flex-2.6.3-3mtl7j/bin/flex (found suitable version "2.6.3", minimum required is "2.6")
-- Found Readline: /usr/include
-- Found MPI_C: /usr/lib64/libpthread.so (found version "3.1")
-- Found MPI_CXX: /usr/lib64/libpthread.so (found version "3.1")
-- Found MPI: TRUE (found version "3.1")
MPT: libxmpi.so 'HPE HMPT 2.25 10/22/21 03:19:55'
-- Found PythonInterp: /gpfs/bbp.cscs.ch/ssd/apps/bsd/2022-01-10/stage_externals/install_gcc-11.2.0-skylake/python-3.9.7-yj5alh/bin/python3 (found suitable version "3.9.7", minimum required is "3.8")
-- Found PythonInterp: /gpfs/bbp.cscs.ch/ssd/apps/bsd/2022-01-10/stage_externals/install_gcc-11.2.0-skylake/python-3.9.7-yj5alh/bin/python3 (found suitable version "3.9.7", minimum required is "3")
-- Found PythonLibs: /gpfs/bbp.cscs.ch/ssd/apps/bsd/2022-01-10/stage_externals/install_gcc-11.2.0-skylake/python-3.9.7-yj5alh/lib/libpython3.9.so
-- Looking for pthread.h
-- Looking for pthread.h - found
-- Performing Test CMAKE_HAVE_LIBC_PTHREAD
-- Performing Test CMAKE_HAVE_LIBC_PTHREAD - Success
-- Found Threads: TRUE
-- Building CoreNEURON
-- 3rd party project: using CLI11 from "external/CLI11"
-- Found Git: /usr/bin/git (found version "1.8.3.1")
-- Setting default CUDA architectures to 70;80
-- The CUDA compiler identification is NVIDIA 11.8.89
-- Detecting CUDA compiler ABI info
-- Detecting CUDA compiler ABI info - done
-- Check for working CUDA compiler: /gpfs/bbp.cscs.ch/ssd/apps/bsd/pulls/1654/stage_externals/install_gcc-11.2.0-skylake/cuda-11.8.0-wlynz6/bin/nvcc - skipped
-- Detecting CUDA compile features
-- Detecting CUDA compile features - done
-- Found CUDAToolkit: /gpfs/bbp.cscs.ch/ssd/apps/bsd/pulls/1654/stage_externals/install_gcc-11.2.0-skylake/cuda-11.8.0-wlynz6/include (found suitable version "11.8.89", minimum required is "9.0")
-- Found Perl: /usr/bin/perl (found version "5.16.3")
-- Found MOD2C: /gpfs/bbp.cscs.ch/home/kumbhar/nrn/external/mod2c
-- Using mod2c submodule from /gpfs/bbp.cscs.ch/home/kumbhar/nrn/external/mod2c
-- mod2c is used as APPLICATION_NAME
-- Found FLEX: /gpfs/bbp.cscs.ch/ssd/apps/bsd/2022-01-10/stage_externals/install_gcc-11.2.0-skylake/flex-2.6.3-3mtl7j/bin/flex (found version "2.6.3")
--
-- CoreNEURON is enabled with following build configuration:
-- --------------------+--------------------------------------------------------
-- Build option | Status
-- --------------------+--------------------------------------------------------
-- CXX COMPILER | /gpfs/bbp.cscs.ch/ssd/apps/bsd/pulls/1654/stage_externals/install_gcc-11.2.0-skylake/nvhpc-22.11-rx7dzj/Linux_x86_64/22.11/compilers/bin/nvc++
-- COMPILE FLAGS | -mp -g -O2 --c++17 -cuda -gpu=cuda11.8,lineinfo,cc70,cc80 -acc -Mautoinline -DCORENEURON_CUDA_PROFILING -DCORENEURON_ENABLE_GPU -DCORENEURON_BUILD -DHAVE_MALLOC_H -DEIGEN_DONT_VECTORIZE=1 -DNRNMPI=1 -DMPI_NO_CPPBIND=1 -DOMPI_SKIP_MPICXX=1 -DMPICH_SKIP_MPICXX=1 -DLAYOUT=0 -DDISABLE_HOC_EXP -DENABLE_SPLAYTREE_QUEUING -DCORENEURON_USE_LEGACY_UNITS=0
-- Build Type | SHARED
-- MPI | ON
-- DYNAMIC | OFF
-- INC | /gpfs/bbp.cscs.ch/ssd/apps/bsd/2022-01-10/stage_externals/install_gcc-11.2.0-skylake/hpe-mpi-2.25.hmpt-4ukyxt/include
-- OpenMP | ON
-- Use legacy units | OFF
-- NMODL | OFF
-- MOD2CPP PATH | /gpfs/bbp.cscs.ch/home/kumbhar/nrn/build/bin/mod2c_core
-- GPU Support | ON
-- CUDA | /gpfs/bbp.cscs.ch/ssd/apps/bsd/pulls/1654/stage_externals/install_gcc-11.2.0-skylake/cuda-11.8.0-wlynz6/lib64
-- Offload | OpenACC
-- Unified Memory | OFF
-- Auto Timeout | ON
-- Wrap exp() | OFF
-- SplayTree Queue | ON
-- NetReceive Buffer | ON
-- Caliper | OFF
-- Likwid | OFF
-- Unit Tests | OFF
-- Reporting | OFF
-- --------------+--------------------------------------------------------------
--
Extracting link flags from target 'Threads::Threads', beware that this can be fragile. Got:
Extracting link flags from path '/usr/lib64/libreadline.so', beware that this can be fragile. Got: -lreadline
Extracting link flags from path '/usr/lib64/libcurses.so', beware that this can be fragile. Got: -lcurses
Extracting link flags from path '/usr/lib64/libform.so', beware that this can be fragile. Got: -lform
Generating link flags from path /gpfs/bbp.cscs.ch/ssd/apps/bsd/2022-01-10/stage_externals/install_gcc-11.2.0-skylake/python-3.9.7-yj5alh/lib/libpython3.9.so Got: /gpfs/bbp.cscs.ch/ssd/apps/bsd/2022-01-10/stage_externals/install_gcc-11.2.0-skylake/python-3.9.7-yj5alh/lib/libpython3.9.so -Wl,-rpath,/gpfs/bbp.cscs.ch/ssd/apps/bsd/2022-01-10/stage_externals/install_gcc-11.2.0-skylake/python-3.9.7-yj5alh/lib
Extracting link flags from path '/usr/lib64/libpthread.so', beware that this can be fragile. Got: -lpthread
Generating link flags from path /gpfs/bbp.cscs.ch/ssd/apps/bsd/2022-01-10/stage_externals/install_gcc-11.2.0-skylake/hpe-mpi-2.25.hmpt-4ukyxt/lib/libmpi.so Got: /gpfs/bbp.cscs.ch/ssd/apps/bsd/2022-01-10/stage_externals/install_gcc-11.2.0-skylake/hpe-mpi-2.25.hmpt-4ukyxt/lib/libmpi.so -Wl,-rpath,/gpfs/bbp.cscs.ch/ssd/apps/bsd/2022-01-10/stage_externals/install_gcc-11.2.0-skylake/hpe-mpi-2.25.hmpt-4ukyxt/lib
Generating link flags from name 'dl', beware that this can be fragile. Got: -ldl
--
-- Configured NEURON 9.0.0
--
-- You can now build NEURON using:
-- cmake --build . --parallel 8 [--target TARGET]
-- You might want to adjust the number of parallel build jobs for your system.
-- Some non-default targets you might want to build:
-- --------------+--------------------------------------------------------------
-- Target | Description
-- --------------+--------------------------------------------------------------
-- install | Will install NEURON to: /gpfs/bbp.cscs.ch/home/kumbhar/install
-- | Change the install location of NEURON using:
-- | cmake <src_path> -DCMAKE_INSTALL_PREFIX=<install_path>
-- docs | Build full docs. Calls targets: doxygen, notebooks, sphinx, notebooks-clean
-- uninstall | Removes files installed by make install (todo)
-- --------------+--------------------------------------------------------------
-- Build option | Status
-- --------------+--------------------------------------------------------------
-- C COMPILER | /gpfs/bbp.cscs.ch/ssd/apps/bsd/pulls/1654/stage_externals/install_gcc-11.2.0-skylake/nvhpc-22.11-rx7dzj/Linux_x86_64/22.11/compilers/bin/nvc
-- CXX COMPILER | /gpfs/bbp.cscs.ch/ssd/apps/bsd/pulls/1654/stage_externals/install_gcc-11.2.0-skylake/nvhpc-22.11-rx7dzj/Linux_x86_64/22.11/compilers/bin/nvc++
-- BUILD_TYPE | RelWithDebInfo (allowed: Custom;Debug;Release;RelWithDebInfo;Fast)
-- COMPILE FLAGS | -g -O2 --diag_suppress=1,47,111,128,170,174,177,180,186,301,541,550,816,941,2465
-- Shared | ON
-- Default units | modern units (2019 nist constants)
-- MPI | ON
-- DYNAMIC | OFF
-- INC | /gpfs/bbp.cscs.ch/ssd/apps/bsd/2022-01-10/stage_externals/install_gcc-11.2.0-skylake/hpe-mpi-2.25.hmpt-4ukyxt/include
-- LIB | /usr/lib64/libpthread.so
-- Python | ON
-- EXE | /gpfs/bbp.cscs.ch/ssd/apps/bsd/2022-01-10/stage_externals/install_gcc-11.2.0-skylake/python-3.9.7-yj5alh/bin/python3
-- INC | /gpfs/bbp.cscs.ch/ssd/apps/bsd/2022-01-10/stage_externals/install_gcc-11.2.0-skylake/python-3.9.7-yj5alh/include/python3.9
-- LIB | /gpfs/bbp.cscs.ch/ssd/apps/bsd/2022-01-10/stage_externals/install_gcc-11.2.0-skylake/python-3.9.7-yj5alh/lib/libpython3.9.so
-- MODULE | ON
-- DYNAMIC | OFF
-- Readline | /usr/lib64/libreadline.so
-- Curses | /usr/lib64/libcurses.so;/usr/lib64/libform.so
-- RX3D | OFF
-- Interviews | OFF
-- CoreNEURON | ON
-- PATH | /gpfs/bbp.cscs.ch/home/kumbhar/nrn/src/coreneuron
-- LINK FLAGS | -cuda -gpu=cuda11.8,lineinfo,cc70,cc80 -acc -lcorenrnmech -lcoreneuron-cuda -lpthread -Wl,-rpath,/gpfs/bbp.cscs.ch/ssd/apps/bsd/2022-01-10/stage_externals/install_gcc-11.2.0-skylake/hpe-mpi-2.25.hmpt-4ukyxt/lib /gpfs/bbp.cscs.ch/ssd/apps/bsd/2022-01-10/stage_externals/install_gcc-11.2.0-skylake/hpe-mpi-2.25.hmpt-4ukyxt/lib/libmpi.so -ldl
-- Legacy Units| OFF
-- Tests | OFF
-- --------------+--------------------------------------------------------------
-- See documentation : https://www.neuron.yale.edu/neuron/
-- --------------+--------------------------------------------------------------
--
-- Configuring done
-- Generating done
-- Build files have been written to: /gpfs/bbp.cscs.ch/home/kumbhar/nrn/build
kumbhar@r2i3n2:~/nrn/build$ make -j12
...
nvc++-Warning-CUDA_HOME has been deprecated. Please, use NVHPC_CUDA_HOME instead.
[100%] Linking CXX executable ../../bin/nrniv
nvc++-Warning-CUDA_HOME has been deprecated. Please, use NVHPC_CUDA_HOME instead.
[100%] Built target nrniv
nvc++-Warning-CUDA_HOME has been deprecated. Please, use NVHPC_CUDA_HOME instead.
[100%] Built target hoc_module
kumbhar@r2i3n2:~/nrn/build$ make install
[ 0%] Built target CLI11
[ 3%] Built target mod2c_core
@pramodk: Thank you for your reply. I have just tried using --parallel 2
and --parallel 1
but got the same error. I have also deleted and re-created the build directory, and ran the same cmake command you showed. I got mostly the same output, except some warnings after -- 3rd party project: using CLI11 from external/CLI11
, as follows:
...
-- Building CoreNEURON
-- 3rd party project: using CLI11 from "external/CLI11"
-- Found Doxygen: /usr/bin/doxygen (found version "1.9.1") found components: doxygen dot
CMake Warning (dev) at /usr/share/cmake-3.22/Modules/CMakeDependentOption.cmake:84 (message):
Policy CMP0127 is not set: cmake_dependent_option() supports full Condition
Syntax. Run "cmake --help-policy CMP0127" for policy details. Use the
cmake_policy command to set the policy and suppress this warning.
Call Stack (most recent call first):
external/CLI11/CMakeLists.txt:66 (cmake_dependent_option)
This warning is for project developers. Use -Wno-dev to suppress it.
CMake Warning (dev) at /usr/share/cmake-3.22/Modules/CMakeDependentOption.cmake:84 (message):
Policy CMP0127 is not set: cmake_dependent_option() supports full Condition
Syntax. Run "cmake --help-policy CMP0127" for policy details. Use the
cmake_policy command to set the policy and suppress this warning.
Call Stack (most recent call first):
external/CLI11/CMakeLists.txt:70 (cmake_dependent_option)
This warning is for project developers. Use -Wno-dev to suppress it.
CMake Warning (dev) at /usr/share/cmake-3.22/Modules/CMakeDependentOption.cmake:84 (message):
Policy CMP0127 is not set: cmake_dependent_option() supports full Condition
Syntax. Run "cmake --help-policy CMP0127" for policy details. Use the
cmake_policy command to set the policy and suppress this warning.
Call Stack (most recent call first):
external/CLI11/CMakeLists.txt:82 (cmake_dependent_option)
This warning is for project developers. Use -Wno-dev to suppress it.
CMake Warning (dev) at /usr/share/cmake-3.22/Modules/CMakeDependentOption.cmake:84 (message):
Policy CMP0127 is not set: cmake_dependent_option() supports full Condition
Syntax. Run "cmake --help-policy CMP0127" for policy details. Use the
cmake_policy command to set the policy and suppress this warning.
Call Stack (most recent call first):
external/CLI11/CMakeLists.txt:94 (cmake_dependent_option)
This warning is for project developers. Use -Wno-dev to suppress it.
CMake Warning (dev) at /usr/share/cmake-3.22/Modules/CMakeDependentOption.cmake:84 (message):
Policy CMP0127 is not set: cmake_dependent_option() supports full Condition
Syntax. Run "cmake --help-policy CMP0127" for policy details. Use the
cmake_policy command to set the policy and suppress this warning.
Call Stack (most recent call first):
external/CLI11/CMakeLists.txt:102 (cmake_dependent_option)
This warning is for project developers. Use -Wno-dev to suppress it.
-- Found Git: /usr/bin/git (found version "2.34.1")
-- Setting default CUDA architectures to 70;80
...
I am attaching the whole CMake log at the end of this post too in case it helps. Let me know if there is any further information I can provide or anything else I can try on my end.
CMake log:
$ cmake .. -DNRN_ENABLE_CORENEURON=ON -DCORENRN_ENABLE_GPU=ON -DNRN_ENABLE_INTERVIEWS=OFF -DNRN_ENABLE_RX3D=OFF -DCMAKE_INSTALL_PREFIX=$HOME/install -DCMAKE_C_COMPILER=nvc -DCMAKE_CXX_COMPILER=nvc++ -DCMAKE_CUDA_COMPILER=nvcc
-- The C compiler identification is NVHPC 22.11.0
-- The CXX compiler identification is NVHPC 22.11.0
-- Detecting C compiler ABI info
-- Detecting C compiler ABI info - done
-- Check for working C compiler: /opt/nvidia/hpc_sdk/Linux_x86_64/22.11/compilers/bin/nvc - skipped
-- Detecting C compile features
-- Detecting C compile features - done
-- Detecting CXX compiler ABI info
-- Detecting CXX compiler ABI info - done
-- Check for working CXX compiler: /opt/nvidia/hpc_sdk/Linux_x86_64/22.11/compilers/bin/nvc++ - skipped
-- Detecting CXX compile features
-- Detecting CXX compile features - done
-- Setting build type to 'RelWithDebInfo' as none was specified.
-- 3rd party project: using Random123 from "external/Random123"
-- -DPYTHON_EXECUTABLE not specified. Looking for `python3` in the PATH exclusively...
-- Setting PYTHON_EXECUTABLE=/usr/bin/python3
-- Found BISON: /usr/bin/bison (found version "3.8.2")
-- Found FLEX: /usr/bin/flex (found suitable version "2.6.4", minimum required is "2.6")
-- Found Readline: /usr/include
-- Found MPI_C: /opt/nvidia/hpc_sdk/Linux_x86_64/22.11/comm_libs/openmpi/openmpi-3.1.5/lib/libmpi.so (found version "3.1")
-- Found MPI_CXX: /opt/nvidia/hpc_sdk/Linux_x86_64/22.11/comm_libs/openmpi/openmpi-3.1.5/lib/libmpi_cxx.so (found version "3.1")
-- Found MPI: TRUE (found version "3.1")
-- Detected OpenMPI 3.1.5
-- Found PythonInterp: /usr/bin/python3 (found suitable version "3.10.6", minimum required is "3.8")
-- Found PythonInterp: /usr/bin/python3 (found suitable version "3.10.6", minimum required is "3")
-- Found PythonLibs: /usr/lib/x86_64-linux-gnu/libpython3.10.so
-- Looking for pthread.h
-- Looking for pthread.h - found
-- Performing Test CMAKE_HAVE_LIBC_PTHREAD
-- Performing Test CMAKE_HAVE_LIBC_PTHREAD - Success
-- Found Threads: TRUE
-- Building CoreNEURON
-- 3rd party project: using CLI11 from "external/CLI11"
-- Found Doxygen: /usr/bin/doxygen (found version "1.9.1") found components: doxygen dot
CMake Warning (dev) at /usr/share/cmake-3.22/Modules/CMakeDependentOption.cmake:84 (message):
Policy CMP0127 is not set: cmake_dependent_option() supports full Condition
Syntax. Run "cmake --help-policy CMP0127" for policy details. Use the
cmake_policy command to set the policy and suppress this warning.
Call Stack (most recent call first):
external/CLI11/CMakeLists.txt:66 (cmake_dependent_option)
This warning is for project developers. Use -Wno-dev to suppress it.
CMake Warning (dev) at /usr/share/cmake-3.22/Modules/CMakeDependentOption.cmake:84 (message):
Policy CMP0127 is not set: cmake_dependent_option() supports full Condition
Syntax. Run "cmake --help-policy CMP0127" for policy details. Use the
cmake_policy command to set the policy and suppress this warning.
Call Stack (most recent call first):
external/CLI11/CMakeLists.txt:70 (cmake_dependent_option)
This warning is for project developers. Use -Wno-dev to suppress it.
CMake Warning (dev) at /usr/share/cmake-3.22/Modules/CMakeDependentOption.cmake:84 (message):
Policy CMP0127 is not set: cmake_dependent_option() supports full Condition
Syntax. Run "cmake --help-policy CMP0127" for policy details. Use the
cmake_policy command to set the policy and suppress this warning.
Call Stack (most recent call first):
external/CLI11/CMakeLists.txt:82 (cmake_dependent_option)
This warning is for project developers. Use -Wno-dev to suppress it.
CMake Warning (dev) at /usr/share/cmake-3.22/Modules/CMakeDependentOption.cmake:84 (message):
Policy CMP0127 is not set: cmake_dependent_option() supports full Condition
Syntax. Run "cmake --help-policy CMP0127" for policy details. Use the
cmake_policy command to set the policy and suppress this warning.
Call Stack (most recent call first):
external/CLI11/CMakeLists.txt:94 (cmake_dependent_option)
This warning is for project developers. Use -Wno-dev to suppress it.
CMake Warning (dev) at /usr/share/cmake-3.22/Modules/CMakeDependentOption.cmake:84 (message):
Policy CMP0127 is not set: cmake_dependent_option() supports full Condition
Syntax. Run "cmake --help-policy CMP0127" for policy details. Use the
cmake_policy command to set the policy and suppress this warning.
Call Stack (most recent call first):
external/CLI11/CMakeLists.txt:102 (cmake_dependent_option)
This warning is for project developers. Use -Wno-dev to suppress it.
-- Found Git: /usr/bin/git (found version "2.34.1")
-- Setting default CUDA architectures to 70;80
-- The CUDA compiler identification is NVIDIA 11.8.89
-- Detecting CUDA compiler ABI info
-- Detecting CUDA compiler ABI info - done
-- Check for working CUDA compiler: /opt/nvidia/hpc_sdk/Linux_x86_64/22.11/compilers/bin/nvcc - skipped
-- Detecting CUDA compile features
-- Detecting CUDA compile features - done
-- Found CUDAToolkit: /opt/nvidia/hpc_sdk/Linux_x86_64/22.11/cuda/11.8/include (found suitable version "11.8.89", minimum required is "9.0")
-- Found Perl: /usr/bin/perl (found version "5.34.0")
-- Found MOD2C: /home/mafalda/nrn/external/mod2c
-- Using mod2c submodule from /home/mafalda/nrn/external/mod2c
-- mod2c is used as APPLICATION_NAME
-- Found FLEX: /usr/bin/flex (found version "2.6.4")
--
-- CoreNEURON is enabled with following build configuration:
-- --------------------+--------------------------------------------------------
-- Build option | Status
-- --------------------+--------------------------------------------------------
-- CXX COMPILER | /opt/nvidia/hpc_sdk/Linux_x86_64/22.11/compilers/bin/nvc++
-- COMPILE FLAGS | -g -O2 --c++17 -cuda -gpu=cuda11.8,lineinfo,cc70,cc80 -acc -Mautoinline -DCORENEURON_CUDA_PROFILING -DCORENEURON_ENABLE_GPU -DCORENEURON_BUILD -DHAVE_MALLOC_H -DEIGEN_DONT_VECTORIZE=1 -DNRNMPI=1 -DMPI_NO_CPPBIND=1 -DOMPI_SKIP_MPICXX=1 -DMPICH_SKIP_MPICXX=1 -DLAYOUT=0 -DDISABLE_HOC_EXP -DENABLE_SPLAYTREE_QUEUING -DCORENEURON_USE_LEGACY_UNITS=0
-- Build Type | SHARED
-- MPI | ON
-- DYNAMIC | OFF
-- INC | /opt/nvidia/hpc_sdk/Linux_x86_64/22.11/comm_libs/openmpi/openmpi-3.1.5/include
-- OpenMP | ON
-- Use legacy units | OFF
-- NMODL | OFF
-- MOD2CPP PATH | /home/mafalda/nrn/build/bin/mod2c_core
-- GPU Support | ON
-- CUDA | /opt/nvidia/hpc_sdk/Linux_x86_64/22.11/cuda/11.8/lib64
-- Offload | OpenACC
-- Unified Memory | OFF
-- Auto Timeout | ON
-- Wrap exp() | OFF
-- SplayTree Queue | ON
-- NetReceive Buffer | ON
-- Caliper | OFF
-- Likwid | OFF
-- Unit Tests | OFF
-- Reporting | OFF
-- --------------+--------------------------------------------------------------
--
Extracting link flags from target 'Threads::Threads', beware that this can be fragile. Got:
Generating link flags from path /usr/lib/x86_64-linux-gnu/libreadline.so Got: /usr/lib/x86_64-linux-gnu/libreadline.so -Wl,-rpath,/usr/lib/x86_64-linux-gnu
Generating link flags from path /usr/lib/x86_64-linux-gnu/libcurses.so Got: /usr/lib/x86_64-linux-gnu/libcurses.so -Wl,-rpath,/usr/lib/x86_64-linux-gnu
Generating link flags from path /usr/lib/x86_64-linux-gnu/libform.so Got: /usr/lib/x86_64-linux-gnu/libform.so -Wl,-rpath,/usr/lib/x86_64-linux-gnu
Generating link flags from path /usr/lib/x86_64-linux-gnu/libpython3.10.so Got: /usr/lib/x86_64-linux-gnu/libpython3.10.so -Wl,-rpath,/usr/lib/x86_64-linux-gnu
Generating link flags from path /opt/nvidia/hpc_sdk/Linux_x86_64/22.11/comm_libs/openmpi/openmpi-3.1.5/lib/libmpi.so Got: /opt/nvidia/hpc_sdk/Linux_x86_64/22.11/comm_libs/openmpi/openmpi-3.1.5/lib/libmpi.so -Wl,-rpath,/opt/nvidia/hpc_sdk/Linux_x86_64/22.11/comm_libs/openmpi/openmpi-3.1.5/lib
Generating link flags from name 'dl', beware that this can be fragile. Got: -ldl
--
-- Configured NEURON 9.0.0
--
-- You can now build NEURON using:
-- cmake --build . --parallel 8 [--target TARGET]
-- You might want to adjust the number of parallel build jobs for your system.
-- Some non-default targets you might want to build:
-- --------------+--------------------------------------------------------------
-- Target | Description
-- --------------+--------------------------------------------------------------
-- install | Will install NEURON to: /home/mafalda/install
-- | Change the install location of NEURON using:
-- | cmake <src_path> -DCMAKE_INSTALL_PREFIX=<install_path>
-- docs | Build full docs. Calls targets: doxygen, notebooks, sphinx, notebooks-clean
-- uninstall | Removes files installed by make install (todo)
-- --------------+--------------------------------------------------------------
-- Build option | Status
-- --------------+--------------------------------------------------------------
-- C COMPILER | /opt/nvidia/hpc_sdk/Linux_x86_64/22.11/compilers/bin/nvc
-- CXX COMPILER | /opt/nvidia/hpc_sdk/Linux_x86_64/22.11/compilers/bin/nvc++
-- BUILD_TYPE | RelWithDebInfo (allowed: Custom;Debug;Release;RelWithDebInfo;Fast)
-- COMPILE FLAGS | -g -O2 --diag_suppress=1,47,111,128,170,174,177,180,186,301,541,550,816,941,2465
-- Shared | ON
-- Default units | modern units (2019 nist constants)
-- MPI | ON
-- DYNAMIC | OFF
-- INC | /opt/nvidia/hpc_sdk/Linux_x86_64/22.11/comm_libs/openmpi/openmpi-3.1.5/include
-- LIB | /opt/nvidia/hpc_sdk/Linux_x86_64/22.11/comm_libs/openmpi/openmpi-3.1.5/lib/libmpi_cxx.so
-- Python | ON
-- EXE | /usr/bin/python3
-- INC | /usr/include/python3.10
-- LIB | /usr/lib/x86_64-linux-gnu/libpython3.10.so
-- MODULE | ON
-- DYNAMIC | OFF
-- Readline | /usr/lib/x86_64-linux-gnu/libreadline.so
-- Curses | /usr/lib/x86_64-linux-gnu/libcurses.so;/usr/lib/x86_64-linux-gnu/libform.so
-- RX3D | OFF
-- Interviews | OFF
-- CoreNEURON | ON
-- PATH | /home/mafalda/nrn/src/coreneuron
-- LINK FLAGS | -cuda -gpu=cuda11.8,lineinfo,cc70,cc80 -acc -lcorenrnmech -lcoreneuron-cuda -Wl,-rpath,/opt/nvidia/hpc_sdk/Linux_x86_64/22.11/comm_libs/openmpi/openmpi-3.1.5/lib /opt/nvidia/hpc_sdk/Linux_x86_64/22.11/comm_libs/openmpi/openmpi-3.1.5/lib/libmpi_cxx.so -Wl,-rpath,/opt/nvidia/hpc_sdk/Linux_x86_64/22.11/comm_libs/openmpi/openmpi-3.1.5/lib /opt/nvidia/hpc_sdk/Linux_x86_64/22.11/comm_libs/openmpi/openmpi-3.1.5/lib/libmpi.so -ldl
-- Legacy Units| OFF
-- Tests | OFF
-- --------------+--------------------------------------------------------------
-- See documentation : https://www.neuron.yale.edu/neuron/
-- --------------+--------------------------------------------------------------
--
-- Configuring done
-- Generating done
-- Build files have been written to: /home/mafalda/nrn/build
Sorry for delay in response.
Not sure how I end-up in this situation but I am seeing same issue on our machine! ("similar" config I reported in previous comment was working).
So just to have the details for the record here, I am adding following information:
On BB5:
kumbhar@r2i3n2:~/workarena/systems/bbpv/repos/bbp/nrn/build_gpu$ module list
Currently Loaded Modulefiles:
1) unstable 3) /flex/2.6.3 5) /python/3.9.7 7) python-dev/0.4 9) /nvhpc/22.11
2) gcc/11.2.0 4) /bison/3.8.2 6) /hpe-mpi/2.25.hmpt 8) /cmake/3.21.4 10) /cuda/11.8.0
$ git rev-parse --short HEAD
c8174d5
[ 4%] Building CXX object src/nrniv/CMakeFiles/nrniv_lib.dir/__/ivoc/ivocvect.cpp.o
cd /gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/build_gpu/src/nrniv && /gpfs/bbp.cscs.ch/ssd/apps/bsd/pulls/1654/stage_externals/install_gcc-11.2.0-skylake/nvhpc-22.11-rx7dzj/Linux_x86_64/22.11/compilers/bin/nvc++ -DHAVE_CONFIG_H -DMPICH_SKIP_MPICXX=1 -DMPI_NO_CPPBIND=1 -DNRN_ENABLE_THREADS -DOMPI_SKIP_MPICXX=1 -DR123_USE_INTRIN_H=0 -Dnrniv_lib_EXPORTS -DUSEBBS=1 -DUSEGNU=1 -DUSEMATRIX=1 -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/ivoc -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/nrncvode -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/nrniv -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/nrnoc -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/oc -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/build_gpu -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/build_gpu/src/nrncvode -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/build_gpu/src/nrnoc -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/build_gpu/src/nrnpython -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/build_gpu/src/oc -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/build_gpu/src/parallel -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/build_gpu/src/sundials -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/build_gpu/src/sundials/shared -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/external/Random123/include -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/gnu -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/mesch -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/nrnmpi -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/nrnpython -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/parallel -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/scopmath -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/sparse -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/sparse13 -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/sundials -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/sundials/cvodes -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/sundials/ida -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/sundials/shared -I/gpfs/bbp.cscs.ch/ssd/apps/bsd/2022-01-10/stage_externals/install_gcc-11.2.0-skylake/python-3.9.7-yj5alh/include/python3.9 -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/build_gpu/src/nrniv/oc_generated -I/gpfs/bbp.cscs.ch/ssd/apps/bsd/2022-01-10/stage_externals/install_gcc-11.2.0-skylake/hpe-mpi-2.25.hmpt-4ukyxt/include -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/nrniv/nrniv_lib -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/ivos -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/build_gpu/src/ivos -g -O2 -fPIC --diag_suppress=1,47,111,128,170,174,177,180,186,301,541,550,816,941,2465 --c++17 -MD -MT src/nrniv/CMakeFiles/nrniv_lib.dir/__/ivoc/ivocvect.cpp.o -MF CMakeFiles/nrniv_lib.dir/__/ivoc/ivocvect.cpp.o.d -o CMakeFiles/nrniv_lib.dir/__/ivoc/ivocvect.cpp.o -c /gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/ivoc/ivocvect.cpp
nvc++-Warning-CUDA_HOME has been deprecated. Please, use NVHPC_CUDA_HOME instead.
nvc++-Fatal-/gpfs/bbp.cscs.ch/ssd/apps/bsd/pulls/1654/stage_externals/install_gcc-11.2.0-skylake/nvhpc-22.11-rx7dzj/Linux_x86_64/22.11/compilers/bin/tools/cpp2 TERMINATED by signal 11
Arguments to /gpfs/bbp.cscs.ch/ssd/apps/bsd/pulls/1654/stage_externals/install_gcc-11.2.0-skylake/nvhpc-22.11-rx7dzj/Linux_x86_64/22.11/compilers/bin/tools/cpp2
/gpfs/bbp.cscs.ch/ssd/apps/bsd/pulls/1654/stage_externals/install_gcc-11.2.0-skylake/nvhpc-22.11-rx7dzj/Linux_x86_64/22.11/compilers/bin/tools/cpp2 /gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/ivoc/ivocvect.cpp -debug -x 120 0x8000 -opt 2 -terse 1 -inform warn -x 119 0xa10000 -x 122 0x40 -x 123 0x1000 -x 127 4 -x 127 17 -x 19 0x400000 -x 28 0x40000 -x 120 0x10000000 -x 70 0x8000 -x 122 1 -x 125 0x20000 -quad -vect 56 -y 34 16 -x 37 0x480000 -x 34 0x8 -x 32 25952256 -y 19 8 -y 35 0 -x 42 0x30 -x 39 0x40 -x 42 0x20000 -x 39 0x80 -x 59 4 -x 129 2 -y 129 0x8000 -tp skylake-avx512 -astype 0 -x 121 1 -fn /gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/ivoc/ivocvect.cpp -il /gpfs/bbp.cscs.ch/ssd/slurmTmpFS/kumbhar/1204718/nvc++CDRYu2Pl5vPh.il -x 117 0x200 -x 123 0x80000000 -x 123 4 -x 119 0x20 -def __pgnu_vsn=110200 -x 70 0x40000000 -x 183 4 -x 121 0x800 -x 6 0x20000 -autoinl 10 -x 168 400 -x 174 128000 -x 14 0x200000 -x 14 0x400000 -autoinl 10 -x 168 400 -x 174 128000 -x 14 0x200000 -x 14 0x400000 -x 249 140 -x 120 0x200000 -x 70 0x40000000 -x 8 0x40000000 -x 164 0x800000 -x 71 0x2000 -x 71 0x4000 -x 34 0x40000000 -x 83 0x1 -x 85 0x1 -x 15 0x1000000 -x 206 0x02 -x 120 0x1000000 -x 68 0x1 -x 39 4 -x 56 0x10 -x 26 0x10 -x 26 1 -x 56 0x4000 -x 197 0 -x 175 0 -x 203 0 -x 204 0 -y 163 0xc0000000 -x 192 0x40000000 -x 189 0x10 -y 189 0x4000000 -x 60 512 -x 9 1 -x 72 0x1 -x 136 0x11 -x 9 1 -x 72 0x1 -x 136 0x11 -x 62 8 -gnuvsn 110200 -x 69 0x200 -x 123 0x400 -x 119 0x08 -cmdline '+nvc++ /gpfs/bbp.cscs.ch/ssd/slurmTmpFS/kumbhar/1204718/nvc++CDRYu2Pl5vPh.il -D__GCC_ATOMIC_TEST_AND_SET_TRUEVAL=1 -DHAVE_CONFIG_H -DMPICH_SKIP_MPICXX=1 -DMPI_NO_CPPBIND=1 -DNRN_ENABLE_THREADS -DOMPI_SKIP_MPICXX=1 -DR123_USE_INTRIN_H=0 -Dnrniv_lib_EXPORTS -DUSEBBS=1 -DUSEGNU=1 -DUSEMATRIX=1 -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/ivoc -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/nrncvode -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/nrniv -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/nrnoc -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/oc -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/build_gpu -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/build_gpu/src/nrncvode -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/build_gpu/src/nrnoc -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/build_gpu/src/nrnpython -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/build_gpu/src/oc -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/build_gpu/src/parallel -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/build_gpu/src/sundials -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/build_gpu/src/sundials/shared -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/external/Random123/include -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/gnu -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/mesch -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/nrnmpi -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/nrnpython -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/parallel -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/scopmath -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/sparse -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/sparse13 -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/sundials -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/sundials/cvodes -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/sundials/ida -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/sundials/shared -I/gpfs/bbp.cscs.ch/ssd/apps/bsd/2022-01-10/stage_externals/install_gcc-11.2.0-skylake/python-3.9.7-yj5alh/include/python3.9 -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/build_gpu/src/nrniv/oc_generated -I/gpfs/bbp.cscs.ch/ssd/apps/bsd/2022-01-10/stage_externals/install_gcc-11.2.0-skylake/hpe-mpi-2.25.hmpt-4ukyxt/include -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/nrniv/nrniv_lib -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/src/ivos -I/gpfs/bbp.cscs.ch/home/kumbhar/workarena/systems/bbpv/repos/bbp/nrn/build_gpu/src/ivos -g -O2 -Mvect=simd -Mno-signed-zeros -fPIC --diag_suppress=1,47,111,128,170,174,177,180,186,301,541,550,816,941,2465 --c++17 -MD -MT src/nrniv/CMakeFiles/nrniv_lib.dir/__/ivoc/ivocvect.cpp.o -MF CMakeFiles/nrniv_lib.dir/__/ivoc/ivocvect.cpp.o.d -o CMakeFiles/nrniv_lib.dir/__/ivoc/ivocvect.cpp.o -c' -asm /gpfs/bbp.cscs.ch/ssd/slurmTmpFS/kumbhar/1204718/nvc++CDRYuV9lwXTy.ll
I will take a look into this bit later today.
@mjribeiro : I can confirm that there is an issue in the latest neuron master with the newer NVHPC compilers. I have reported compiler bug to NVIDIA/NVHPC forum here: https://forums.developer.nvidia.com/t/cpp2-terminated-by-signal-11-error-with-newer-nvhpc-releases-like-23-1-22-11/242001.
The error is triggered by the last week's change in NEURON: https://github.com/neuronsimulator/nrn/pull/2200.
Until we find a workaround, one option I can suggest is to use an older commit than the above mentioned change i.e. if you have cloned git repo then do:
git checkout 2ad762b934b2ffc42b21b3d609fd10091535ea22
git submodule update -f --init --recursive # not strictly required as the moment as submodules are not changed
and then build the project same way.
Let me know if that allows you to build the process successfully.
@pramodk Thank you for your suggestion. I tried using the older commit with NVHPC 22.11 but was still getting the same errors, so I ended up reverting to NVHPC 22.9 as well and it seems to be installing correctly now. Thanks for your help with this!
Describe the issue I am currently trying to install NEURON/CoreNEURON following the instructions on the README in this repo, but the build and install step is not working. See logs and system specs below. Thank you!
To Reproduce
-DCORENRN_ENABLE_GPU=ON
for step 3/4cmake --build . --parallel 8 --target install
Expected behavior The build and install step should not fail
Logs Output from
cmake --build . --parallel 8 --target install
:System