Error in installing pytorch 1.1.0 with cuda 9.0 from source

1

[1208/4410] Building CXX object third_p…dir/src/FbgemmFloat16ConvertAvx512.cc.
FAILED: third_party/fbgemm/CMakeFiles/fbgemm_avx512.dir/src/FbgemmFloat16ConvertAvx512.cc.o
/opt/rh/devtoolset-4/root/usr/bin/c++ -DFBGEMM_STATIC -DTH_BLAS_MKL -I…/third_party/cpuinfo/include -I…/third_party/fbgemm/third_party/asmjit/src -I…/third_party/fbgemm/include -I…/third_party/fbgemm -I…/cmake/…/third_party/benchmark/include -isystem …/cmake/…/third_party/googletest/googlemock/include -isystem …/cmake/…/third_party/googletest/googletest/include -isystem …/third_party/protobuf/src -isystem /home/liuxiyang/anaconda3/envs/torch1_1/include -isystem …/third_party/gemmlowp -isystem …/third_party/neon2sse -isystem …/third_party/XNNPACK/include -Wno-deprecated -fvisibility-inlines-hidden -fopenmp -O3 -DNDEBUG -fPIC -fvisibility=hidden -m64 -mavx2 -mfma -mavx512f -mavx512bw -mavx512dq -mavx512vl -masm=intel -std=c++11 -MD -MT third_party/fbgemm/CMakeFiles/fbgemm_avx512.dir/src/FbgemmFloat16ConvertAvx512.cc.o -MF third_party/fbgemm/CMakeFiles/fbgemm_avx512.dir/src/FbgemmFloat16ConvertAvx512.cc.o.d -o third_party/fbgemm/CMakeFiles/fbgemm_avx512.dir/src/FbgemmFloat16ConvertAvx512.cc.o -c …/third_party/fbgemm/src/FbgemmFloat16ConvertAvx512.cc
/tmp/ccp0wFOC.s: Assembler messages:
/tmp/ccp0wFOC.s:53: Error: operand size mismatch for vbroadcastss' /tmp/ccp0wFOC.s:55: Error: operand size mismatch for vbroadcastss’
[1227/4410] Building CXX object third_p…avx2.dir/src/FbgemmI8DepthwiseAvx2.cc.
ninja: build stopped: subcommand failed.
Traceback (most recent call last):
File “setup.py”, line 737, in
build_deps()
File “setup.py”, line 316, in build_deps
cmake=cmake)
File “/home/liuxiyang/peige/DFDC/pytorch/tools/build_pytorch_libs.py”, line 62, in build_caffe2
cmake.build(my_env)
File “/home/liuxiyang/peige/DFDC/pytorch/tools/setup_helpers/cmake.py”, line 339, in build
self.run(build_args, my_env)
File “/home/liuxiyang/peige/DFDC/pytorch/tools/setup_helpers/cmake.py”, line 141, in run
check_call(command, cwd=self.build_dir, env=env)
File “/home/liuxiyang/anaconda3/envs/torch1_1/lib/python3.6/subprocess.py”, line 311, in check_call
raise CalledProcessError(retcode, cmd)
subprocess.CalledProcessError: Command ‘[‘cmake’, ‘–build’, ‘.’, ‘–target’, ‘install’, ‘–config’, ‘Release’, ‘–’, ‘-j’, ‘20’]’ returned non-zero exit status 1.

What should I do to solve this problem? Thank you.

g++ version is 5.3.1
cmake version is 3.14.0

2

Could you try to sync the submodules and rerun the installation?

git submodule sync && git submodule update --init --recursive
3

Hi, I met the same problem and have synced the submodules, but the problem still exists, are there any other solutions.

4

Could you try to uninstall all PyTorch installations, sync the submodules and rebuild?
Also, if that doesn’t help, create a new (conda) environment and try to build PyTorch there.

If that still doesn’t help, could you please post the stack trace here?

5

I updated my python to 3.7 and then used the following command to solve this problem.

git submodule sync && git submodule update --init --recursive

------------------ Original ------------------