Description
Cannot fully compile mxnet on ubuntu because of out of memory error.
Environment info (Required)
Linux ip-172-31-12-26 4.4.0-1048-aws #57-Ubuntu SMP Tue Jan 9 21:45:57 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux
Package used (Python/R/Scala/Julia):
Python
Build info (Required if built from source)
Compiler gcc.
MXNet commit hash:
This is my requirement.
HEAD detached at 998378a
Error Message:
cc1plus: out of memory allocating 85573424 bytes after a total of 383340544 bytes
Makefile:270: recipe for target 'build/src/operator/tensor/control_flow_op.o' failed
make: *** [build/src/operator/tensor/control_flow_op.o] Error 1
Minimum reproducible example
make -j $(nproc) USE_OPENCV=1 USE_BLAS=openblas
Steps to reproduce
https://mxnet.incubator.apache.org/get_started/install.html
Linux/Python/CPU/Build from Sources
Step 4: $ make -j $(nproc) USE_OPENCV=1 USE_BLAS=openblas
command:make -j $(nproc) USE_OPENCV=1 USE_BLAS=openblas
Output:
g++ -std=c++11 -c -DMSHADOW_FORCE_STREAM -Wall -Wsign-compare -O3 -DNDEBUG=1 -I/home/ubuntu/mxnet/mshadow/ -I/home/ubuntu/mxnet/dmlc-core/include -fPIC -I/home/ubuntu/mxnet/nnvm/include -I/home/ubuntu/mxnet/dlpack/include -Iinclude -funroll-loops -Wno-unused-variable -Wno-unused-parameter -Wno-unknown-pragmas -Wno-unused-local-typedefs -msse3 -DMSHADOW_USE_CUDA=0 -DMSHADOW_USE_CBLAS=1 -DMSHADOW_USE_MKL=0 -DMSHADOW_RABIT_PS=0 -DMSHADOW_DIST_PS=0 -DMSHADOW_USE_PASCAL=0 -DMXNET_USE_OPENCV=1 -I/usr/include/opencv -fopenmp -DMXNET_USE_LAPACK -DMXNET_USE_NVRTC=0 -MMD -c src/operator/tensor/control_flow_op.cc -o build/src/operator/tensor/control_flow_op.o
cc1plus: out of memory allocating 85573424 bytes after a total of 383340544 bytes
Makefile:270: recipe for target 'build/src/operator/tensor/control_flow_op.o' failed
make: *** [build/src/operator/tensor/control_flow_op.o] Error 1
What have you tried to solve it?
Could those files be created with different encoding on Windows, which can cause my problem?
I'm not sure about this and I'm not sure what is the best way to deal with it. Can I have not enough RAM?
I tried without flag -j too.
Description
Cannot fully compile mxnet on ubuntu because of out of memory error.
Environment info (Required)
Linux ip-172-31-12-26 4.4.0-1048-aws #57-Ubuntu SMP Tue Jan 9 21:45:57 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux
Package used (Python/R/Scala/Julia):
Python
Build info (Required if built from source)
Compiler gcc.
MXNet commit hash:
This is my requirement.
HEAD detached at 998378a
Error Message:
cc1plus: out of memory allocating 85573424 bytes after a total of 383340544 bytes
Makefile:270: recipe for target 'build/src/operator/tensor/control_flow_op.o' failed
make: *** [build/src/operator/tensor/control_flow_op.o] Error 1
Minimum reproducible example
make -j $(nproc) USE_OPENCV=1 USE_BLAS=openblas
Steps to reproduce
https://mxnet.incubator.apache.org/get_started/install.html
Linux/Python/CPU/Build from Sources
Step 4: $ make -j $(nproc) USE_OPENCV=1 USE_BLAS=openblas
command:make -j $(nproc) USE_OPENCV=1 USE_BLAS=openblas
Output:
g++ -std=c++11 -c -DMSHADOW_FORCE_STREAM -Wall -Wsign-compare -O3 -DNDEBUG=1 -I/home/ubuntu/mxnet/mshadow/ -I/home/ubuntu/mxnet/dmlc-core/include -fPIC -I/home/ubuntu/mxnet/nnvm/include -I/home/ubuntu/mxnet/dlpack/include -Iinclude -funroll-loops -Wno-unused-variable -Wno-unused-parameter -Wno-unknown-pragmas -Wno-unused-local-typedefs -msse3 -DMSHADOW_USE_CUDA=0 -DMSHADOW_USE_CBLAS=1 -DMSHADOW_USE_MKL=0 -DMSHADOW_RABIT_PS=0 -DMSHADOW_DIST_PS=0 -DMSHADOW_USE_PASCAL=0 -DMXNET_USE_OPENCV=1 -I/usr/include/opencv -fopenmp -DMXNET_USE_LAPACK -DMXNET_USE_NVRTC=0 -MMD -c src/operator/tensor/control_flow_op.cc -o build/src/operator/tensor/control_flow_op.o
cc1plus: out of memory allocating 85573424 bytes after a total of 383340544 bytes
Makefile:270: recipe for target 'build/src/operator/tensor/control_flow_op.o' failed
make: *** [build/src/operator/tensor/control_flow_op.o] Error 1
What have you tried to solve it?
Could those files be created with different encoding on Windows, which can cause my problem?
I'm not sure about this and I'm not sure what is the best way to deal with it. Can I have not enough RAM?
I tried without flag -j too.