search

tusen-ai / simpledet

A Simple and Versatile Framework for Object Detection and Instance Recognition
Apache License 2.0
3.08k stars 488 forks source link

Make error when building from scratch on Ubuntu 18.04 #169

Closed loki900718 closed 5 years ago

loki900718 commented 5 years ago

Thanks for the awesome work.

Following the 'Setup from Scratch' guide, I was able to complete installing the prerequisites and setting up the environmental variables. But I met the following error trying to build the mxnet source:

>> make -j$((`nproc`-1))
Makefile:345: WARNING: Significant performance increases can be achieved by installing and enabling gperftools or jemalloc development packages
INFO: nvcc was not found on your path
INFO: Using /usr/local/cuda/bin/nvcc as nvcc path
Running CUDA_ARCH: -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=sm_50 -gencode arch=compute_60,code=sm_60 -gencode arch=compute_70,code=sm_70
make CXX="g++" DEPS_PATH="/tmp/mxnet/deps" -C /tmp/mxnet/3rdparty/ps-lite ps
cd /tmp/mxnet/3rdparty/dmlc-core; make libdmlc.a USE_SSE=1 config=/tmp/mxnet/config.mk; cd /tmp/mxnet
make[1]: Entering directory '/tmp/mxnet/3rdparty/dmlc-core'
make[1]: Entering directory '/tmp/mxnet/3rdparty/ps-lite'
/tmp/mxnet/deps/bin/protoc --cpp_out=./src --proto_path=./src src/meta.proto
make[1]: 'libdmlc.a' is up to date.
make[1]: Leaving directory '/tmp/mxnet/3rdparty/dmlc-core'
/tmp/mxnet/deps/bin/protoc --cpp_out=./src --proto_path=./src src/./meta.proto
g++ -I./src -I./include -I/tmp/mxnet/deps/include -std=c++0x -MM -MT build/postoffice.o src/postoffice.cc >build/postoffice.d
g++ -I./src -I./include -I/tmp/mxnet/deps/include -std=c++0x -MM -MT build/van.o src/van.cc >build/van.d
g++ -I./src -I./include -I/tmp/mxnet/deps/include -std=c++0x -MM -MT build/customer.o src/customer.cc >build/customer.d
g++ -I./src -I./include -I/tmp/mxnet/deps/include -std=c++0x -MM -MT build/meta.pb.o src/meta.pb.cc >build/meta.pb.d
g++ -std=c++11 -msse2 -fPIC -O3 -ggdb -Wall -finline-functions -I./src -I./include -I/tmp/mxnet/deps/include  -c src/customer.cc -o build/customer.o
g++ -std=c++11 -msse2 -fPIC -O3 -ggdb -Wall -finline-functions -I./src -I./include -I/tmp/mxnet/deps/include  -c src/postoffice.cc -o build/postoffice.o
g++ -std=c++11 -msse2 -fPIC -O3 -ggdb -Wall -finline-functions -I./src -I./include -I/tmp/mxnet/deps/include  -c src/meta.pb.cc -o build/meta.pb.o
g++ -std=c++11 -msse2 -fPIC -O3 -ggdb -Wall -finline-functions -I./src -I./include -I/tmp/mxnet/deps/include  -c src/van.cc -o build/van.o
/usr/local/cuda/bin/nvcc -std=c++11 -Xcompiler -D_FORCE_INLINES -O3 -ccbin g++  -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=sm_50 -gencode arch=compute_60,code=sm_60 -gencode arch=compute_70,code=sm_70 -Xcompiler "-DMSHADOW_FORCE_STREAM -Wall -Wsign-compare -O3 -DNDEBUG=1 -I/tmp/mxnet/3rdparty/mshadow/ -I/tmp/mxnet/3rdparty/dmlc-core/include -fPIC -I/tmp/mxnet/3rdparty/tvm/nnvm/include -I/tmp/mxnet/3rdparty/dlpack/include -I/tmp/mxnet/3rdparty/tvm/include -Iinclude -funroll-loops -Wno-unused-parameter -Wno-unknown-pragmas -Wno-unused-local-typedefs -msse3 -mf16c -I/usr/local/cuda/include -DMSHADOW_USE_CBLAS=1 -DMSHADOW_USE_MKL=0 -DMSHADOW_RABIT_PS=0 -DMSHADOW_DIST_PS=0 -DMSHADOW_USE_PASCAL=0 -DMXNET_USE_SIGNAL_HANDLER=1 -DMXNET_USE_OPENCV=0 -DMSHADOW_INT64_TENSOR_SIZE=0 -DMSHADOW_USE_CUDNN=1  -DMXNET_USE_DIST_KVSTORE -I/tmp/mxnet/3rdparty/ps-lite/include -I/tmp/mxnet/deps/include -I/tmp/mxnet/3rdparty/nvidia_cub -I/include -DMXNET_USE_NCCL=1 -DMXNET_USE_LIBJPEG_TURBO=0" --generate-dependencies -MT build/src/operator/contrib/group_norm_helper_gpu.o src/operator/contrib/group_norm_helper.cu >build/src/operator/contrib/group_norm_helper_gpu.d
/usr/local/cuda/bin/nvcc -std=c++11 -Xcompiler -D_FORCE_INLINES -O3 -ccbin g++  -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=sm_50 -gencode arch=compute_60,code=sm_60 -gencode arch=compute_70,code=sm_70 -Xcompiler "-DMSHADOW_FORCE_STREAM -Wall -Wsign-compare -O3 -DNDEBUG=1 -I/tmp/mxnet/3rdparty/mshadow/ -I/tmp/mxnet/3rdparty/dmlc-core/include -fPIC -I/tmp/mxnet/3rdparty/tvm/nnvm/include -I/tmp/mxnet/3rdparty/dlpack/include -I/tmp/mxnet/3rdparty/tvm/include -Iinclude -funroll-loops -Wno-unused-parameter -Wno-unknown-pragmas -Wno-unused-local-typedefs -msse3 -mf16c -I/usr/local/cuda/include -DMSHADOW_USE_CBLAS=1 -DMSHADOW_USE_MKL=0 -DMSHADOW_RABIT_PS=0 -DMSHADOW_DIST_PS=0 -DMSHADOW_USE_PASCAL=0 -DMXNET_USE_SIGNAL_HANDLER=1 -DMXNET_USE_OPENCV=0 -DMSHADOW_INT64_TENSOR_SIZE=0 -DMSHADOW_USE_CUDNN=1  -DMXNET_USE_DIST_KVSTORE -I/tmp/mxnet/3rdparty/ps-lite/include -I/tmp/mxnet/deps/include -I/tmp/mxnet/3rdparty/nvidia_cub -I/include -DMXNET_USE_NCCL=1 -DMXNET_USE_LIBJPEG_TURBO=0" --generate-dependencies -MT build/src/operator/contrib/multibox_prior_gpu.o src/operator/contrib/multibox_prior.cu >build/src/operator/contrib/multibox_prior_gpu.d
/usr/local/cuda/bin/nvcc -std=c++11 -Xcompiler -D_FORCE_INLINES -O3 -ccbin g++  -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=sm_50 -gencode arch=compute_60,code=sm_60 -gencode arch=compute_70,code=sm_70 -Xcompiler "-DMSHADOW_FORCE_STREAM -Wall -Wsign-compare -O3 -DNDEBUG=1 -I/tmp/mxnet/3rdparty/mshadow/ -I/tmp/mxnet/3rdparty/dmlc-core/include -fPIC -I/tmp/mxnet/3rdparty/tvm/nnvm/include -I/tmp/mxnet/3rdparty/dlpack/include -I/tmp/mxnet/3rdparty/tvm/include -Iinclude -funroll-loops -Wno-unused-parameter -Wno-unknown-pragmas -Wno-unused-local-typedefs -msse3 -mf16c -I/usr/local/cuda/include -DMSHADOW_USE_CBLAS=1 -DMSHADOW_USE_MKL=0 -DMSHADOW_RABIT_PS=0 -DMSHADOW_DIST_PS=0 -DMSHADOW_USE_PASCAL=0 -DMXNET_USE_SIGNAL_HANDLER=1 -DMXNET_USE_OPENCV=0 -DMSHADOW_INT64_TENSOR_SIZE=0 -DMSHADOW_USE_CUDNN=1  -DMXNET_USE_DIST_KVSTORE -I/tmp/mxnet/3rdparty/ps-lite/include -I/tmp/mxnet/deps/include -I/tmp/mxnet/3rdparty/nvidia_cub -I/include -DMXNET_USE_NCCL=1 -DMXNET_USE_LIBJPEG_TURBO=0" --generate-dependencies -MT build/src/operator/contrib/all_finite_gpu.o src/operator/contrib/all_finite.cu >build/src/operator/contrib/all_finite_gpu.d
/usr/local/cuda/bin/nvcc -std=c++11 -Xcompiler -D_FORCE_INLINES -O3 -ccbin g++  -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=sm_50 -gencode arch=compute_60,code=sm_60 -gencode arch=compute_70,code=sm_70 -Xcompiler "-DMSHADOW_FORCE_STREAM -Wall -Wsign-compare -O3 -DNDEBUG=1 -I/tmp/mxnet/3rdparty/mshadow/ -I/tmp/mxnet/3rdparty/dmlc-core/include -fPIC -I/tmp/mxnet/3rdparty/tvm/nnvm/include -I/tmp/mxnet/3rdparty/dlpack/include -I/tmp/mxnet/3rdparty/tvm/include -Iinclude -funroll-loops -Wno-unused-parameter -Wno-unknown-pragmas -Wno-unused-local-typedefs -msse3 -mf16c -I/usr/local/cuda/include -DMSHADOW_USE_CBLAS=1 -DMSHADOW_USE_MKL=0 -DMSHADOW_RABIT_PS=0 -DMSHADOW_DIST_PS=0 -DMSHADOW_USE_PASCAL=0 -DMXNET_USE_SIGNAL_HANDLER=1 -DMXNET_USE_OPENCV=0 -DMSHADOW_INT64_TENSOR_SIZE=0 -DMSHADOW_USE_CUDNN=1  -DMXNET_USE_DIST_KVSTORE -I/tmp/mxnet/3rdparty/ps-lite/include -I/tmp/mxnet/deps/include -I/tmp/mxnet/3rdparty/nvidia_cub -I/include -DMXNET_USE_NCCL=1 -DMXNET_USE_LIBJPEG_TURBO=0" --generate-dependencies -MT build/src/operator/contrib/hawkes_ll_gpu.o src/operator/contrib/hawkes_ll.cu >build/src/operator/contrib/hawkes_ll_gpu.d
/usr/local/cuda/bin/nvcc -std=c++11 -Xcompiler -D_FORCE_INLINES -O3 -ccbin g++  -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=sm_50 -gencode arch=compute_60,code=sm_60 -gencode arch=compute_70,code=sm_70 -Xcompiler "-DMSHADOW_FORCE_STREAM -Wall -Wsign-compare -O3 -DNDEBUG=1 -I/tmp/mxnet/3rdparty/mshadow/ -I/tmp/mxnet/3rdparty/dmlc-core/include -fPIC -I/tmp/mxnet/3rdparty/tvm/nnvm/include -I/tmp/mxnet/3rdparty/dlpack/include -I/tmp/mxnet/3rdparty/tvm/include -Iinclude -funroll-loops -Wno-unused-parameter -Wno-unknown-pragmas -Wno-unused-local-typedefs -msse3 -mf16c -I/usr/local/cuda/include -DMSHADOW_USE_CBLAS=1 -DMSHADOW_USE_MKL=0 -DMSHADOW_RABIT_PS=0 -DMSHADOW_DIST_PS=0 -DMSHADOW_USE_PASCAL=0 -DMXNET_USE_SIGNAL_HANDLER=1 -DMXNET_USE_OPENCV=0 -DMSHADOW_INT64_TENSOR_SIZE=0 -DMSHADOW_USE_CUDNN=1  -DMXNET_USE_DIST_KVSTORE -I/tmp/mxnet/3rdparty/ps-lite/include -I/tmp/mxnet/deps/include -I/tmp/mxnet/3rdparty/nvidia_cub -I/include -DMXNET_USE_NCCL=1 -DMXNET_USE_LIBJPEG_TURBO=0" --generate-dependencies -MT build/src/operator/contrib/roi_align_v2_gpu.o src/operator/contrib/roi_align_v2.cu >build/src/operator/contrib/roi_align_v2_gpu.d
/usr/local/cuda/bin/nvcc -std=c++11 -Xcompiler -D_FORCE_INLINES -O3 -ccbin g++  -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=sm_50 -gencode arch=compute_60,code=sm_60 -gencode arch=compute_70,code=sm_70 -Xcompiler "-DMSHADOW_FORCE_STREAM -Wall -Wsign-compare -O3 -DNDEBUG=1 -I/tmp/mxnet/3rdparty/mshadow/ -I/tmp/mxnet/3rdparty/dmlc-core/include -fPIC -I/tmp/mxnet/3rdparty/tvm/nnvm/include -I/tmp/mxnet/3rdparty/dlpack/include -I/tmp/mxnet/3rdparty/tvm/include -Iinclude -funroll-loops -Wno-unused-parameter -Wno-unknown-pragmas -Wno-unused-local-typedefs -msse3 -mf16c -I/usr/local/cuda/include -DMSHADOW_USE_CBLAS=1 -DMSHADOW_USE_MKL=0 -DMSHADOW_RABIT_PS=0 -DMSHADOW_DIST_PS=0 -DMSHADOW_USE_PASCAL=0 -DMXNET_USE_SIGNAL_HANDLER=1 -DMXNET_USE_OPENCV=0 -DMSHADOW_INT64_TENSOR_SIZE=0 -DMSHADOW_USE_CUDNN=1  -DMXNET_USE_DIST_KVSTORE -I/tmp/mxnet/3rdparty/ps-lite/include -I/tmp/mxnet/deps/include -I/tmp/mxnet/3rdparty/nvidia_cub -I/include -DMXNET_USE_NCCL=1 -DMXNET_USE_LIBJPEG_TURBO=0" --generate-dependencies -MT build/src/operator/contrib/transformer_gpu.o src/operator/contrib/transformer.cu >build/src/operator/contrib/transformer_gpu.d
/usr/local/cuda/bin/nvcc -std=c++11 -Xcompiler -D_FORCE_INLINES -O3 -ccbin g++  -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=sm_50 -gencode arch=compute_60,code=sm_60 -gencode arch=compute_70,code=sm_70 -Xcompiler "-DMSHADOW_FORCE_STREAM -Wall -Wsign-compare -O3 -DNDEBUG=1 -I/tmp/mxnet/3rdparty/mshadow/ -I/tmp/mxnet/3rdparty/dmlc-core/include -fPIC -I/tmp/mxnet/3rdparty/tvm/nnvm/include -I/tmp/mxnet/3rdparty/dlpack/include -I/tmp/mxnet/3rdparty/tvm/include -Iinclude -funroll-loops -Wno-unused-parameter -Wno-unknown-pragmas -Wno-unused-local-typedefs -msse3 -mf16c -I/usr/local/cuda/include -DMSHADOW_USE_CBLAS=1 -DMSHADOW_USE_MKL=0 -DMSHADOW_RABIT_PS=0 -DMSHADOW_DIST_PS=0 -DMSHADOW_USE_PASCAL=0 -DMXNET_USE_SIGNAL_HANDLER=1 -DMXNET_USE_OPENCV=0 -DMSHADOW_INT64_TENSOR_SIZE=0 -DMSHADOW_USE_CUDNN=1  -DMXNET_USE_DIST_KVSTORE -I/tmp/mxnet/3rdparty/ps-lite/include -I/tmp/mxnet/deps/include -I/tmp/mxnet/3rdparty/nvidia_cub -I/include -DMXNET_USE_NCCL=1 -DMXNET_USE_LIBJPEG_TURBO=0" --generate-dependencies -MT build/src/operator/custom/native_op_gpu.o src/operator/custom/native_op.cu >build/src/operator/custom/native_op_gpu.d
/usr/local/cuda/bin/nvcc -std=c++11 -Xcompiler -D_FORCE_INLINES -O3 -ccbin g++  -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=sm_50 -gencode arch=compute_60,code=sm_60 -gencode arch=compute_70,code=sm_70 -Xcompiler "-DMSHADOW_FORCE_STREAM -Wall -Wsign-compare -O3 -DNDEBUG=1 -I/tmp/mxnet/3rdparty/mshadow/ -I/tmp/mxnet/3rdparty/dmlc-core/include -fPIC -I/tmp/mxnet/3rdparty/tvm/nnvm/include -I/tmp/mxnet/3rdparty/dlpack/include -I/tmp/mxnet/3rdparty/tvm/include -Iinclude -funroll-loops -Wno-unused-parameter -Wno-unknown-pragmas -Wno-unused-local-typedefs -msse3 -mf16c -I/usr/local/cuda/include -DMSHADOW_USE_CBLAS=1 -DMSHADOW_USE_MKL=0 -DMSHADOW_RABIT_PS=0 -DMSHADOW_DIST_PS=0 -DMSHADOW_USE_PASCAL=0 -DMXNET_USE_SIGNAL_HANDLER=1 -DMXNET_USE_OPENCV=0 -DMSHADOW_INT64_TENSOR_SIZE=0 -DMSHADOW_USE_CUDNN=1  -DMXNET_USE_DIST_KVSTORE -I/tmp/mxnet/3rdparty/ps-lite/include -I/tmp/mxnet/deps/include -I/tmp/mxnet/3rdparty/nvidia_cub -I/include -DMXNET_USE_NCCL=1 -DMXNET_USE_LIBJPEG_TURBO=0" --generate-dependencies -MT build/src/common/utils_gpu.o src/common/utils.cu >build/src/common/utils_gpu.d
/usr/local/cuda/bin/nvcc -std=c++11 -Xcompiler -D_FORCE_INLINES -O3 -ccbin g++  -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=sm_50 -gencode arch=compute_60,code=sm_60 -gencode arch=compute_70,code=sm_70 -Xcompiler "-DMSHADOW_FORCE_STREAM -Wall -Wsign-compare -O3 -DNDEBUG=1 -I/tmp/mxnet/3rdparty/mshadow/ -I/tmp/mxnet/3rdparty/dmlc-core/include -fPIC -I/tmp/mxnet/3rdparty/tvm/nnvm/include -I/tmp/mxnet/3rdparty/dlpack/include -I/tmp/mxnet/3rdparty/tvm/include -Iinclude -funroll-loops -Wno-unused-parameter -Wno-unknown-pragmas -Wno-unused-local-typedefs -msse3 -mf16c -I/usr/local/cuda/include -DMSHADOW_USE_CBLAS=1 -DMSHADOW_USE_MKL=0 -DMSHADOW_RABIT_PS=0 -DMSHADOW_DIST_PS=0 -DMSHADOW_USE_PASCAL=0 -DMXNET_USE_SIGNAL_HANDLER=1 -DMXNET_USE_OPENCV=0 -DMSHADOW_INT64_TENSOR_SIZE=0 -DMSHADOW_USE_CUDNN=1  -DMXNET_USE_DIST_KVSTORE -I/tmp/mxnet/3rdparty/ps-lite/include -I/tmp/mxnet/deps/include -I/tmp/mxnet/3rdparty/nvidia_cub -I/include -DMXNET_USE_NCCL=1 -DMXNET_USE_LIBJPEG_TURBO=0" --generate-dependencies -MT build/src/common/random_generator_gpu.o src/common/random_generator.cu >build/src/common/random_generator_gpu.d
/usr/local/cuda/bin/nvcc -std=c++11 -Xcompiler -D_FORCE_INLINES -O3 -ccbin g++  -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=sm_50 -gencode arch=compute_60,code=sm_60 -gencode arch=compute_70,code=sm_70 -Xcompiler "-DMSHADOW_FORCE_STREAM -Wall -Wsign-compare -O3 -DNDEBUG=1 -I/tmp/mxnet/3rdparty/mshadow/ -I/tmp/mxnet/3rdparty/dmlc-core/include -fPIC -I/tmp/mxnet/3rdparty/tvm/nnvm/include -I/tmp/mxnet/3rdparty/dlpack/include -I/tmp/mxnet/3rdparty/tvm/include -Iinclude -funroll-loops -Wno-unused-parameter -Wno-unknown-pragmas -Wno-unused-local-typedefs -msse3 -mf16c -I/usr/local/cuda/include -DMSHADOW_USE_CBLAS=1 -DMSHADOW_USE_MKL=0 -DMSHADOW_RABIT_PS=0 -DMSHADOW_DIST_PS=0 -DMSHADOW_USE_PASCAL=0 -DMXNET_USE_SIGNAL_HANDLER=1 -DMXNET_USE_OPENCV=0 -DMSHADOW_INT64_TENSOR_SIZE=0 -DMSHADOW_USE_CUDNN=1  -DMXNET_USE_DIST_KVSTORE -I/tmp/mxnet/3rdparty/ps-lite/include -I/tmp/mxnet/deps/include -I/tmp/mxnet/3rdparty/nvidia_cub -I/include -DMXNET_USE_NCCL=1 -DMXNET_USE_LIBJPEG_TURBO=0" --generate-dependencies -MT build/src/ndarray/ndarray_function_gpu.o src/ndarray/ndarray_function.cu >build/src/ndarray/ndarray_function_gpu.d
/usr/local/cuda/bin/nvcc -std=c++11 -Xcompiler -D_FORCE_INLINES -O3 -ccbin g++  -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=sm_50 -gencode arch=compute_60,code=sm_60 -gencode arch=compute_70,code=sm_70 -Xcompiler "-DMSHADOW_FORCE_STREAM -Wall -Wsign-compare -O3 -DNDEBUG=1 -I/tmp/mxnet/3rdparty/mshadow/ -I/tmp/mxnet/3rdparty/dmlc-core/include -fPIC -I/tmp/mxnet/3rdparty/tvm/nnvm/include -I/tmp/mxnet/3rdparty/dlpack/include -I/tmp/mxnet/3rdparty/tvm/include -Iinclude -funroll-loops -Wno-unused-parameter -Wno-unknown-pragmas -Wno-unused-local-typedefs -msse3 -mf16c -I/usr/local/cuda/include -DMSHADOW_USE_CBLAS=1 -DMSHADOW_USE_MKL=0 -DMSHADOW_RABIT_PS=0 -DMSHADOW_DIST_PS=0 -DMSHADOW_USE_PASCAL=0 -DMXNET_USE_SIGNAL_HANDLER=1 -DMXNET_USE_OPENCV=0 -DMSHADOW_INT64_TENSOR_SIZE=0 -DMSHADOW_USE_CUDNN=1  -DMXNET_USE_DIST_KVSTORE -I/tmp/mxnet/3rdparty/ps-lite/include -I/tmp/mxnet/deps/include -I/tmp/mxnet/3rdparty/nvidia_cub -I/include -DMXNET_USE_NCCL=1 -DMXNET_USE_LIBJPEG_TURBO=0" --generate-dependencies -MT build/src/operator/svm_output_gpu.o src/operator/svm_output.cu >build/src/operator/svm_output_gpu.d
src/postoffice.cc: In member function ‘void ps::Postoffice::Manage(const ps::Message&)’:
src/postoffice.cc:186:43: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
     for (int customer_id = 0; customer_id < barrier_done_[recv.meta.app_id].size();
                               ~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
src/van.cc: In member function ‘void ps::Van::UpdateLocalID(ps::Message*, std::unordered_set<int>*, ps::Meta*, ps::Meta*)’:
src/van.cc:118:36: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
     if (nodes->control.node.size() < num_nodes) {
         ~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~
/usr/local/cuda/bin/nvcc -c -o build/src/operator/contrib/group_norm_helper_gpu.o -std=c++11 -Xcompiler -D_FORCE_INLINES -O3 -ccbin g++  -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=sm_50 -gencode arch=compute_60,code=sm_60 -gencode arch=compute_70,code=sm_70 -Xcompiler "-DMSHADOW_FORCE_STREAM -Wall -Wsign-compare -O3 -DNDEBUG=1 -I/tmp/mxnet/3rdparty/mshadow/ -I/tmp/mxnet/3rdparty/dmlc-core/include -fPIC -I/tmp/mxnet/3rdparty/tvm/nnvm/include -I/tmp/mxnet/3rdparty/dlpack/include -I/tmp/mxnet/3rdparty/tvm/include -Iinclude -funroll-loops -Wno-unused-parameter -Wno-unknown-pragmas -Wno-unused-local-typedefs -msse3 -mf16c -I/usr/local/cuda/include -DMSHADOW_USE_CBLAS=1 -DMSHADOW_USE_MKL=0 -DMSHADOW_RABIT_PS=0 -DMSHADOW_DIST_PS=0 -DMSHADOW_USE_PASCAL=0 -DMXNET_USE_SIGNAL_HANDLER=1 -DMXNET_USE_OPENCV=0 -DMSHADOW_INT64_TENSOR_SIZE=0 -DMSHADOW_USE_CUDNN=1  -DMXNET_USE_DIST_KVSTORE -I/tmp/mxnet/3rdparty/ps-lite/include -I/tmp/mxnet/deps/include -I/tmp/mxnet/3rdparty/nvidia_cub -I/include -DMXNET_USE_NCCL=1 -DMXNET_USE_LIBJPEG_TURBO=0" src/operator/contrib/group_norm_helper.cu
/usr/local/cuda/bin/nvcc -std=c++11 -Xcompiler -D_FORCE_INLINES -O3 -ccbin g++  -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=sm_50 -gencode arch=compute_60,code=sm_60 -gencode arch=compute_70,code=sm_70 -Xcompiler "-DMSHADOW_FORCE_STREAM -Wall -Wsign-compare -O3 -DNDEBUG=1 -I/tmp/mxnet/3rdparty/mshadow/ -I/tmp/mxnet/3rdparty/dmlc-core/include -fPIC -I/tmp/mxnet/3rdparty/tvm/nnvm/include -I/tmp/mxnet/3rdparty/dlpack/include -I/tmp/mxnet/3rdparty/tvm/include -Iinclude -funroll-loops -Wno-unused-parameter -Wno-unknown-pragmas -Wno-unused-local-typedefs -msse3 -mf16c -I/usr/local/cuda/include -DMSHADOW_USE_CBLAS=1 -DMSHADOW_USE_MKL=0 -DMSHADOW_RABIT_PS=0 -DMSHADOW_DIST_PS=0 -DMSHADOW_USE_PASCAL=0 -DMXNET_USE_SIGNAL_HANDLER=1 -DMXNET_USE_OPENCV=0 -DMSHADOW_INT64_TENSOR_SIZE=0 -DMSHADOW_USE_CUDNN=1  -DMXNET_USE_DIST_KVSTORE -I/tmp/mxnet/3rdparty/ps-lite/include -I/tmp/mxnet/deps/include -I/tmp/mxnet/3rdparty/nvidia_cub -I/include -DMXNET_USE_NCCL=1 -DMXNET_USE_LIBJPEG_TURBO=0" --generate-dependencies -MT build/src/operator/optimizer_op_gpu.o src/operator/optimizer_op.cu >build/src/operator/optimizer_op_gpu.d
In file included from ./include/ps/internal/customer.h:14:0,
                 from src/customer.cc:4:
./include/ps/internal/message.h: In destructor ‘ps::Customer::~Customer()’:
./include/ps/internal/message.h:101:8: warning: ‘*((void*)(& msg)+88).ps::Control::barrier_group’ may be used uninitialized in this function [-Wmaybe-uninitialized]
 struct Control {
        ^~~~~~~
./include/ps/internal/message.h:101:8: warning: ‘*((void*)(& msg)+88).ps::Control::msg_sig’ may be used uninitialized in this function [-Wmaybe-uninitialized]
/usr/local/cuda/bin/nvcc -std=c++11 -Xcompiler -D_FORCE_INLINES -O3 -ccbin g++  -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=sm_50 -gencode arch=compute_60,code=sm_60 -gencode arch=compute_70,code=sm_70 -Xcompiler "-DMSHADOW_FORCE_STREAM -Wall -Wsign-compare -O3 -DNDEBUG=1 -I/tmp/mxnet/3rdparty/mshadow/ -I/tmp/mxnet/3rdparty/dmlc-core/include -fPIC -I/tmp/mxnet/3rdparty/tvm/nnvm/include -I/tmp/mxnet/3rdparty/dlpack/include -I/tmp/mxnet/3rdparty/tvm/include -Iinclude -funroll-loops -Wno-unused-parameter -Wno-unknown-pragmas -Wno-unused-local-typedefs -msse3 -mf16c -I/usr/local/cuda/include -DMSHADOW_USE_CBLAS=1 -DMSHADOW_USE_MKL=0 -DMSHADOW_RABIT_PS=0 -DMSHADOW_DIST_PS=0 -DMSHADOW_USE_PASCAL=0 -DMXNET_USE_SIGNAL_HANDLER=1 -DMXNET_USE_OPENCV=0 -DMSHADOW_INT64_TENSOR_SIZE=0 -DMSHADOW_USE_CUDNN=1  -DMXNET_USE_DIST_KVSTORE -I/tmp/mxnet/3rdparty/ps-lite/include -I/tmp/mxnet/deps/include -I/tmp/mxnet/3rdparty/nvidia_cub -I/include -DMXNET_USE_NCCL=1 -DMXNET_USE_LIBJPEG_TURBO=0" --generate-dependencies -MT build/src/operator/pooling_v1_gpu.o src/operator/pooling_v1.cu >build/src/operator/pooling_v1_gpu.d
src/operator/contrib/group_norm_helper.cu(332): error: namespace "std" has no member "accumulate"

src/operator/contrib/group_norm_helper.cu(333): error: namespace "std" has no member "accumulate"

2 errors detected in the compilation of "/tmp/tmpxft_000031ac_00000000-9_group_norm_helper.compute_70.cpp1.ii".
Makefile:535: recipe for target 'build/src/operator/contrib/group_norm_helper_gpu.o' failed
make: *** [build/src/operator/contrib/group_norm_helper_gpu.o] Error 1
make: *** Waiting for unfinished jobs....
/usr/local/cuda/bin/nvcc -c -o build/src/operator/contrib/transformer_gpu.o -std=c++11 -Xcompiler -D_FORCE_INLINES -O3 -ccbin g++  -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=sm_50 -gencode arch=compute_60,code=sm_60 -gencode arch=compute_70,code=sm_70 -Xcompiler "-DMSHADOW_FORCE_STREAM -Wall -Wsign-compare -O3 -DNDEBUG=1 -I/tmp/mxnet/3rdparty/mshadow/ -I/tmp/mxnet/3rdparty/dmlc-core/include -fPIC -I/tmp/mxnet/3rdparty/tvm/nnvm/include -I/tmp/mxnet/3rdparty/dlpack/include -I/tmp/mxnet/3rdparty/tvm/include -Iinclude -funroll-loops -Wno-unused-parameter -Wno-unknown-pragmas -Wno-unused-local-typedefs -msse3 -mf16c -I/usr/local/cuda/include -DMSHADOW_USE_CBLAS=1 -DMSHADOW_USE_MKL=0 -DMSHADOW_RABIT_PS=0 -DMSHADOW_DIST_PS=0 -DMSHADOW_USE_PASCAL=0 -DMXNET_USE_SIGNAL_HANDLER=1 -DMXNET_USE_OPENCV=0 -DMSHADOW_INT64_TENSOR_SIZE=0 -DMSHADOW_USE_CUDNN=1  -DMXNET_USE_DIST_KVSTORE -I/tmp/mxnet/3rdparty/ps-lite/include -I/tmp/mxnet/deps/include -I/tmp/mxnet/3rdparty/nvidia_cub -I/include -DMXNET_USE_NCCL=1 -DMXNET_USE_LIBJPEG_TURBO=0" src/operator/contrib/transformer.cu
/usr/local/cuda/bin/nvcc -c -o build/src/operator/custom/native_op_gpu.o -std=c++11 -Xcompiler -D_FORCE_INLINES -O3 -ccbin g++  -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=sm_50 -gencode arch=compute_60,code=sm_60 -gencode arch=compute_70,code=sm_70 -Xcompiler "-DMSHADOW_FORCE_STREAM -Wall -Wsign-compare -O3 -DNDEBUG=1 -I/tmp/mxnet/3rdparty/mshadow/ -I/tmp/mxnet/3rdparty/dmlc-core/include -fPIC -I/tmp/mxnet/3rdparty/tvm/nnvm/include -I/tmp/mxnet/3rdparty/dlpack/include -I/tmp/mxnet/3rdparty/tvm/include -Iinclude -funroll-loops -Wno-unused-parameter -Wno-unknown-pragmas -Wno-unused-local-typedefs -msse3 -mf16c -I/usr/local/cuda/include -DMSHADOW_USE_CBLAS=1 -DMSHADOW_USE_MKL=0 -DMSHADOW_RABIT_PS=0 -DMSHADOW_DIST_PS=0 -DMSHADOW_USE_PASCAL=0 -DMXNET_USE_SIGNAL_HANDLER=1 -DMXNET_USE_OPENCV=0 -DMSHADOW_INT64_TENSOR_SIZE=0 -DMSHADOW_USE_CUDNN=1  -DMXNET_USE_DIST_KVSTORE -I/tmp/mxnet/3rdparty/ps-lite/include -I/tmp/mxnet/deps/include -I/tmp/mxnet/3rdparty/nvidia_cub -I/include -DMXNET_USE_NCCL=1 -DMXNET_USE_LIBJPEG_TURBO=0" src/operator/custom/native_op.cu
/usr/local/cuda/bin/nvcc -c -o build/src/common/random_generator_gpu.o -std=c++11 -Xcompiler -D_FORCE_INLINES -O3 -ccbin g++  -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=sm_50 -gencode arch=compute_60,code=sm_60 -gencode arch=compute_70,code=sm_70 -Xcompiler "-DMSHADOW_FORCE_STREAM -Wall -Wsign-compare -O3 -DNDEBUG=1 -I/tmp/mxnet/3rdparty/mshadow/ -I/tmp/mxnet/3rdparty/dmlc-core/include -fPIC -I/tmp/mxnet/3rdparty/tvm/nnvm/include -I/tmp/mxnet/3rdparty/dlpack/include -I/tmp/mxnet/3rdparty/tvm/include -Iinclude -funroll-loops -Wno-unused-parameter -Wno-unknown-pragmas -Wno-unused-local-typedefs -msse3 -mf16c -I/usr/local/cuda/include -DMSHADOW_USE_CBLAS=1 -DMSHADOW_USE_MKL=0 -DMSHADOW_RABIT_PS=0 -DMSHADOW_DIST_PS=0 -DMSHADOW_USE_PASCAL=0 -DMXNET_USE_SIGNAL_HANDLER=1 -DMXNET_USE_OPENCV=0 -DMSHADOW_INT64_TENSOR_SIZE=0 -DMSHADOW_USE_CUDNN=1  -DMXNET_USE_DIST_KVSTORE -I/tmp/mxnet/3rdparty/ps-lite/include -I/tmp/mxnet/deps/include -I/tmp/mxnet/3rdparty/nvidia_cub -I/include -DMXNET_USE_NCCL=1 -DMXNET_USE_LIBJPEG_TURBO=0" src/common/random_generator.cu
/usr/local/cuda/bin/nvcc -c -o build/src/operator/contrib/multibox_prior_gpu.o -std=c++11 -Xcompiler -D_FORCE_INLINES -O3 -ccbin g++  -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=sm_50 -gencode arch=compute_60,code=sm_60 -gencode arch=compute_70,code=sm_70 -Xcompiler "-DMSHADOW_FORCE_STREAM -Wall -Wsign-compare -O3 -DNDEBUG=1 -I/tmp/mxnet/3rdparty/mshadow/ -I/tmp/mxnet/3rdparty/dmlc-core/include -fPIC -I/tmp/mxnet/3rdparty/tvm/nnvm/include -I/tmp/mxnet/3rdparty/dlpack/include -I/tmp/mxnet/3rdparty/tvm/include -Iinclude -funroll-loops -Wno-unused-parameter -Wno-unknown-pragmas -Wno-unused-local-typedefs -msse3 -mf16c -I/usr/local/cuda/include -DMSHADOW_USE_CBLAS=1 -DMSHADOW_USE_MKL=0 -DMSHADOW_RABIT_PS=0 -DMSHADOW_DIST_PS=0 -DMSHADOW_USE_PASCAL=0 -DMXNET_USE_SIGNAL_HANDLER=1 -DMXNET_USE_OPENCV=0 -DMSHADOW_INT64_TENSOR_SIZE=0 -DMSHADOW_USE_CUDNN=1  -DMXNET_USE_DIST_KVSTORE -I/tmp/mxnet/3rdparty/ps-lite/include -I/tmp/mxnet/deps/include -I/tmp/mxnet/3rdparty/nvidia_cub -I/include -DMXNET_USE_NCCL=1 -DMXNET_USE_LIBJPEG_TURBO=0" src/operator/contrib/multibox_prior.cu
/usr/local/cuda/bin/nvcc -c -o build/src/operator/contrib/hawkes_ll_gpu.o -std=c++11 -Xcompiler -D_FORCE_INLINES -O3 -ccbin g++  -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=sm_50 -gencode arch=compute_60,code=sm_60 -gencode arch=compute_70,code=sm_70 -Xcompiler "-DMSHADOW_FORCE_STREAM -Wall -Wsign-compare -O3 -DNDEBUG=1 -I/tmp/mxnet/3rdparty/mshadow/ -I/tmp/mxnet/3rdparty/dmlc-core/include -fPIC -I/tmp/mxnet/3rdparty/tvm/nnvm/include -I/tmp/mxnet/3rdparty/dlpack/include -I/tmp/mxnet/3rdparty/tvm/include -Iinclude -funroll-loops -Wno-unused-parameter -Wno-unknown-pragmas -Wno-unused-local-typedefs -msse3 -mf16c -I/usr/local/cuda/include -DMSHADOW_USE_CBLAS=1 -DMSHADOW_USE_MKL=0 -DMSHADOW_RABIT_PS=0 -DMSHADOW_DIST_PS=0 -DMSHADOW_USE_PASCAL=0 -DMXNET_USE_SIGNAL_HANDLER=1 -DMXNET_USE_OPENCV=0 -DMSHADOW_INT64_TENSOR_SIZE=0 -DMSHADOW_USE_CUDNN=1  -DMXNET_USE_DIST_KVSTORE -I/tmp/mxnet/3rdparty/ps-lite/include -I/tmp/mxnet/deps/include -I/tmp/mxnet/3rdparty/nvidia_cub -I/include -DMXNET_USE_NCCL=1 -DMXNET_USE_LIBJPEG_TURBO=0" src/operator/contrib/hawkes_ll.cu
/usr/local/cuda/bin/nvcc -c -o build/src/operator/svm_output_gpu.o -std=c++11 -Xcompiler -D_FORCE_INLINES -O3 -ccbin g++  -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=sm_50 -gencode arch=compute_60,code=sm_60 -gencode arch=compute_70,code=sm_70 -Xcompiler "-DMSHADOW_FORCE_STREAM -Wall -Wsign-compare -O3 -DNDEBUG=1 -I/tmp/mxnet/3rdparty/mshadow/ -I/tmp/mxnet/3rdparty/dmlc-core/include -fPIC -I/tmp/mxnet/3rdparty/tvm/nnvm/include -I/tmp/mxnet/3rdparty/dlpack/include -I/tmp/mxnet/3rdparty/tvm/include -Iinclude -funroll-loops -Wno-unused-parameter -Wno-unknown-pragmas -Wno-unused-local-typedefs -msse3 -mf16c -I/usr/local/cuda/include -DMSHADOW_USE_CBLAS=1 -DMSHADOW_USE_MKL=0 -DMSHADOW_RABIT_PS=0 -DMSHADOW_DIST_PS=0 -DMSHADOW_USE_PASCAL=0 -DMXNET_USE_SIGNAL_HANDLER=1 -DMXNET_USE_OPENCV=0 -DMSHADOW_INT64_TENSOR_SIZE=0 -DMSHADOW_USE_CUDNN=1  -DMXNET_USE_DIST_KVSTORE -I/tmp/mxnet/3rdparty/ps-lite/include -I/tmp/mxnet/deps/include -I/tmp/mxnet/3rdparty/nvidia_cub -I/include -DMXNET_USE_NCCL=1 -DMXNET_USE_LIBJPEG_TURBO=0" src/operator/svm_output.cu
/usr/local/cuda/bin/nvcc -c -o build/src/operator/contrib/all_finite_gpu.o -std=c++11 -Xcompiler -D_FORCE_INLINES -O3 -ccbin g++  -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=sm_50 -gencode arch=compute_60,code=sm_60 -gencode arch=compute_70,code=sm_70 -Xcompiler "-DMSHADOW_FORCE_STREAM -Wall -Wsign-compare -O3 -DNDEBUG=1 -I/tmp/mxnet/3rdparty/mshadow/ -I/tmp/mxnet/3rdparty/dmlc-core/include -fPIC -I/tmp/mxnet/3rdparty/tvm/nnvm/include -I/tmp/mxnet/3rdparty/dlpack/include -I/tmp/mxnet/3rdparty/tvm/include -Iinclude -funroll-loops -Wno-unused-parameter -Wno-unknown-pragmas -Wno-unused-local-typedefs -msse3 -mf16c -I/usr/local/cuda/include -DMSHADOW_USE_CBLAS=1 -DMSHADOW_USE_MKL=0 -DMSHADOW_RABIT_PS=0 -DMSHADOW_DIST_PS=0 -DMSHADOW_USE_PASCAL=0 -DMXNET_USE_SIGNAL_HANDLER=1 -DMXNET_USE_OPENCV=0 -DMSHADOW_INT64_TENSOR_SIZE=0 -DMSHADOW_USE_CUDNN=1  -DMXNET_USE_DIST_KVSTORE -I/tmp/mxnet/3rdparty/ps-lite/include -I/tmp/mxnet/deps/include -I/tmp/mxnet/3rdparty/nvidia_cub -I/include -DMXNET_USE_NCCL=1 -DMXNET_USE_LIBJPEG_TURBO=0" src/operator/contrib/all_finite.cu
/usr/local/cuda/bin/nvcc -c -o build/src/operator/contrib/roi_align_v2_gpu.o -std=c++11 -Xcompiler -D_FORCE_INLINES -O3 -ccbin g++  -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=sm_50 -gencode arch=compute_60,code=sm_60 -gencode arch=compute_70,code=sm_70 -Xcompiler "-DMSHADOW_FORCE_STREAM -Wall -Wsign-compare -O3 -DNDEBUG=1 -I/tmp/mxnet/3rdparty/mshadow/ -I/tmp/mxnet/3rdparty/dmlc-core/include -fPIC -I/tmp/mxnet/3rdparty/tvm/nnvm/include -I/tmp/mxnet/3rdparty/dlpack/include -I/tmp/mxnet/3rdparty/tvm/include -Iinclude -funroll-loops -Wno-unused-parameter -Wno-unknown-pragmas -Wno-unused-local-typedefs -msse3 -mf16c -I/usr/local/cuda/include -DMSHADOW_USE_CBLAS=1 -DMSHADOW_USE_MKL=0 -DMSHADOW_RABIT_PS=0 -DMSHADOW_DIST_PS=0 -DMSHADOW_USE_PASCAL=0 -DMXNET_USE_SIGNAL_HANDLER=1 -DMXNET_USE_OPENCV=0 -DMSHADOW_INT64_TENSOR_SIZE=0 -DMSHADOW_USE_CUDNN=1  -DMXNET_USE_DIST_KVSTORE -I/tmp/mxnet/3rdparty/ps-lite/include -I/tmp/mxnet/deps/include -I/tmp/mxnet/3rdparty/nvidia_cub -I/include -DMXNET_USE_NCCL=1 -DMXNET_USE_LIBJPEG_TURBO=0" src/operator/contrib/roi_align_v2.cu
/usr/local/cuda/bin/nvcc -c -o build/src/common/utils_gpu.o -std=c++11 -Xcompiler -D_FORCE_INLINES -O3 -ccbin g++  -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=sm_50 -gencode arch=compute_60,code=sm_60 -gencode arch=compute_70,code=sm_70 -Xcompiler "-DMSHADOW_FORCE_STREAM -Wall -Wsign-compare -O3 -DNDEBUG=1 -I/tmp/mxnet/3rdparty/mshadow/ -I/tmp/mxnet/3rdparty/dmlc-core/include -fPIC -I/tmp/mxnet/3rdparty/tvm/nnvm/include -I/tmp/mxnet/3rdparty/dlpack/include -I/tmp/mxnet/3rdparty/tvm/include -Iinclude -funroll-loops -Wno-unused-parameter -Wno-unknown-pragmas -Wno-unused-local-typedefs -msse3 -mf16c -I/usr/local/cuda/include -DMSHADOW_USE_CBLAS=1 -DMSHADOW_USE_MKL=0 -DMSHADOW_RABIT_PS=0 -DMSHADOW_DIST_PS=0 -DMSHADOW_USE_PASCAL=0 -DMXNET_USE_SIGNAL_HANDLER=1 -DMXNET_USE_OPENCV=0 -DMSHADOW_INT64_TENSOR_SIZE=0 -DMSHADOW_USE_CUDNN=1  -DMXNET_USE_DIST_KVSTORE -I/tmp/mxnet/3rdparty/ps-lite/include -I/tmp/mxnet/deps/include -I/tmp/mxnet/3rdparty/nvidia_cub -I/include -DMXNET_USE_NCCL=1 -DMXNET_USE_LIBJPEG_TURBO=0" src/common/utils.cu
/usr/local/cuda/bin/nvcc -c -o build/src/ndarray/ndarray_function_gpu.o -std=c++11 -Xcompiler -D_FORCE_INLINES -O3 -ccbin g++  -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=sm_50 -gencode arch=compute_60,code=sm_60 -gencode arch=compute_70,code=sm_70 -Xcompiler "-DMSHADOW_FORCE_STREAM -Wall -Wsign-compare -O3 -DNDEBUG=1 -I/tmp/mxnet/3rdparty/mshadow/ -I/tmp/mxnet/3rdparty/dmlc-core/include -fPIC -I/tmp/mxnet/3rdparty/tvm/nnvm/include -I/tmp/mxnet/3rdparty/dlpack/include -I/tmp/mxnet/3rdparty/tvm/include -Iinclude -funroll-loops -Wno-unused-parameter -Wno-unknown-pragmas -Wno-unused-local-typedefs -msse3 -mf16c -I/usr/local/cuda/include -DMSHADOW_USE_CBLAS=1 -DMSHADOW_USE_MKL=0 -DMSHADOW_RABIT_PS=0 -DMSHADOW_DIST_PS=0 -DMSHADOW_USE_PASCAL=0 -DMXNET_USE_SIGNAL_HANDLER=1 -DMXNET_USE_OPENCV=0 -DMSHADOW_INT64_TENSOR_SIZE=0 -DMSHADOW_USE_CUDNN=1  -DMXNET_USE_DIST_KVSTORE -I/tmp/mxnet/3rdparty/ps-lite/include -I/tmp/mxnet/deps/include -I/tmp/mxnet/3rdparty/nvidia_cub -I/include -DMXNET_USE_NCCL=1 -DMXNET_USE_LIBJPEG_TURBO=0" src/ndarray/ndarray_function.cu
/usr/local/cuda/bin/nvcc -c -o build/src/operator/optimizer_op_gpu.o -std=c++11 -Xcompiler -D_FORCE_INLINES -O3 -ccbin g++  -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=sm_50 -gencode arch=compute_60,code=sm_60 -gencode arch=compute_70,code=sm_70 -Xcompiler "-DMSHADOW_FORCE_STREAM -Wall -Wsign-compare -O3 -DNDEBUG=1 -I/tmp/mxnet/3rdparty/mshadow/ -I/tmp/mxnet/3rdparty/dmlc-core/include -fPIC -I/tmp/mxnet/3rdparty/tvm/nnvm/include -I/tmp/mxnet/3rdparty/dlpack/include -I/tmp/mxnet/3rdparty/tvm/include -Iinclude -funroll-loops -Wno-unused-parameter -Wno-unknown-pragmas -Wno-unused-local-typedefs -msse3 -mf16c -I/usr/local/cuda/include -DMSHADOW_USE_CBLAS=1 -DMSHADOW_USE_MKL=0 -DMSHADOW_RABIT_PS=0 -DMSHADOW_DIST_PS=0 -DMSHADOW_USE_PASCAL=0 -DMXNET_USE_SIGNAL_HANDLER=1 -DMXNET_USE_OPENCV=0 -DMSHADOW_INT64_TENSOR_SIZE=0 -DMSHADOW_USE_CUDNN=1  -DMXNET_USE_DIST_KVSTORE -I/tmp/mxnet/3rdparty/ps-lite/include -I/tmp/mxnet/deps/include -I/tmp/mxnet/3rdparty/nvidia_cub -I/include -DMXNET_USE_NCCL=1 -DMXNET_USE_LIBJPEG_TURBO=0" src/operator/optimizer_op.cu
/usr/local/cuda/bin/nvcc -c -o build/src/operator/pooling_v1_gpu.o -std=c++11 -Xcompiler -D_FORCE_INLINES -O3 -ccbin g++  -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=sm_50 -gencode arch=compute_60,code=sm_60 -gencode arch=compute_70,code=sm_70 -Xcompiler "-DMSHADOW_FORCE_STREAM -Wall -Wsign-compare -O3 -DNDEBUG=1 -I/tmp/mxnet/3rdparty/mshadow/ -I/tmp/mxnet/3rdparty/dmlc-core/include -fPIC -I/tmp/mxnet/3rdparty/tvm/nnvm/include -I/tmp/mxnet/3rdparty/dlpack/include -I/tmp/mxnet/3rdparty/tvm/include -Iinclude -funroll-loops -Wno-unused-parameter -Wno-unknown-pragmas -Wno-unused-local-typedefs -msse3 -mf16c -I/usr/local/cuda/include -DMSHADOW_USE_CBLAS=1 -DMSHADOW_USE_MKL=0 -DMSHADOW_RABIT_PS=0 -DMSHADOW_DIST_PS=0 -DMSHADOW_USE_PASCAL=0 -DMXNET_USE_SIGNAL_HANDLER=1 -DMXNET_USE_OPENCV=0 -DMSHADOW_INT64_TENSOR_SIZE=0 -DMSHADOW_USE_CUDNN=1  -DMXNET_USE_DIST_KVSTORE -I/tmp/mxnet/3rdparty/ps-lite/include -I/tmp/mxnet/deps/include -I/tmp/mxnet/3rdparty/nvidia_cub -I/include -DMXNET_USE_NCCL=1 -DMXNET_USE_LIBJPEG_TURBO=0" src/operator/pooling_v1.cu
ar crv build/libps.a build/customer.o build/postoffice.o build/van.o build/meta.pb.o
r - build/customer.o
r - build/postoffice.o
r - build/van.o
r - build/meta.pb.o
make[1]: Leaving directory '/tmp/mxnet/3rdparty/ps-lite'

It looks like the compilation of the build/src/operator/contrib/group_norm_helper_gpu.o operator has failed for some reason. Don't have much clue on how to debug this one. Here's my environment info: OS: Ubuntu 18.04 Cuda: 10.0 cudnn: 7.5

RogerChern commented 5 years ago

MXNet just added their GN days ago, I will patch the codes later. An immediate workaround is to remove group norm files under src/operator/nn

SimpleXP commented 5 years ago

This method works! But you also need to remove group_norm* files from src/operator/contrib