Compile error with CUDA10.1 and Pytorch 1.4.0

Hello, I met this error when I try to run the command "pip install -v --no-cache-dir --global-option="--cpp_ext" --global-option="--cuda_ext" .":

I only show you the parts at the start and the end because the ignored part are very similar.

Non-user install because site-packages writeable^M Created temporary directory: /tmp/pip-ephem-wheel-cache-35u048z4^M Created temporary directory: /tmp/pip-req-tracker-o_dho7t1^M Initialized build tracking at /tmp/pip-req-tracker-o_dho7t1^M Created build tracker: /tmp/pip-req-tracker-o_dho7t1^M Entered build tracker: /tmp/pip-req-tracker-o_dho7t1^M Created temporary directory: /tmp/pip-install-lcb6uyn6^M Processing /home/guangrui/apex^M Created temporary directory: /tmp/pip-req-build-_dh7ztxj^M Added file:///home/guangrui/apex to build tracker '/tmp/pip-req-tracker-o_dho7t1'^M Running setup.py (path:/tmp/pip-req-build-_dh7ztxj/setup.py) egg_info for package from file:///home/guangrui/apex^M Running command python setup.py egg_info^M torch.__version__ = 1.4.0^M running egg_info^M creating /tmp/pip-req-build-_dh7ztxj/pip-egg-info/apex.egg-info^M writing /tmp/pip-req-build-_dh7ztxj/pip-egg-info/apex.egg-info/PKG-INFO^M writing dependency_links to /tmp/pip-req-build-_dh7ztxj/pip-egg-info/apex.egg-info/dependency_links.txt^M writing top-level names to /tmp/pip-req-build-_dh7ztxj/pip-egg-info/apex.egg-info/top_level.txt^M writing manifest file '/tmp/pip-req-build-_dh7ztxj/pip-egg-info/apex.egg-info/SOURCES.txt'^M reading manifest file '/tmp/pip-req-build-_dh7ztxj/pip-egg-info/apex.egg-info/SOURCES.txt'^M writing manifest file '/tmp/pip-req-build-_dh7ztxj/pip-egg-info/apex.egg-info/SOURCES.txt'^M /tmp/pip-req-build-_dh7ztxj/setup.py:43: UserWarning: Option --pyprof not specified. Not installing PyProf dependencies!^M warnings.warn("Option --pyprof not specified. Not installing PyProf dependencies!")^M Source in /tmp/pip-req-build-_dh7ztxj has version 0.1, which satisfies requirement apex==0.1 from file:///home/guangrui/apex^M Removed apex==0.1 from file:///home/guangrui/apex from build tracker '/tmp/pip-req-tracker-o_dho7t1'^M Skipping wheel build for apex, due to binaries being disabled for it.^M Installing collected packages: apex^M Attempting uninstall: apex^M Found existing installation: apex 0.1^M Uninstalling apex-0.1:^M Created temporary directory: /tmp/pip-uninstall-v72mrjmz^M Removing file or directory /home/guangrui/anaconda3/lib/python3.7/site-packages/apex-0.1-py3.7.egg^M Removing pth entries from /home/guangrui/anaconda3/lib/python3.7/site-packages/easy-install.pth:^M Removing entry: ./apex-0.1-py3.7.egg^M Successfully uninstalled apex-0.1^M Created temporary directory: /tmp/pip-record-gcgz29ft^M Running command /home/guangrui/anaconda3/bin/python -u -c 'import sys, setuptools, tokenize; sys.argv[0] = '"'"'/tmp/pip-req-build-_dh7ztxj/setup.py'"'"'; __file__='"'"'/tmp/pip-req-build-_dh7ztxj/setup.py'"'"';f=getattr(tokenize, '"'"'open'"'"', open)(__file__);code=f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, __file__, '"'"'exec'"'"'))' --cpp_ext --cuda_ext install --record /tmp/pip-record-gcgz29ft/install-record.txt --single-version-externally-managed --compile --install-headers /home/guangrui/anaconda3/include/python3.7m/apex^M torch.__version__ = 1.4.0^M /tmp/pip-req-build-_dh7ztxj/setup.py:43: UserWarning: Option --pyprof not specified. Not installing PyProf dependencies!^M warnings.warn("Option --pyprof not specified. Not installing PyProf dependencies!")^M ^M Compiling cuda extensions with^M nvcc: NVIDIA (R) Cuda compiler driver^M Copyright (c) 2005-2019 NVIDIA Corporation^M Built on Sun_Jul_28_19:07:16_PDT_2019^M Cuda compilation tools, release 10.1, V10.1.243^M from /opt/cuda/bin^M ^M running install^M running build^M onda3/lib/python3.7/site-packages/torch/include/torch/csrc/api/include -I/home/guangrui/anaconda3/lib/python3.7/site-packages/torch/include/TH -I/home/guangrui/anaconda3/lib/python3.7/site-packages/torch/include/THC -I/home/guangrui/anaconda3/include/python3.7m -c csrc/flatten_unflatten.cpp -o build/temp.linux-x86_64-3.7/csrc/flatten_unflatten.o -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_EXTENSION_NAME=apex_C -D_GLIBCXX_USE_CXX11_ABI=0 -std=c++11^M cc1plus: warning: command line option ‘-Wstrict-prototypes’ is valid for C/ObjC but not for C++^M In file included from /home/guangrui/anaconda3/lib/python3.7/site-packages/torch/include/ATen/core/TensorMethods.h:10,^M from /home/guangrui/anaconda3/lib/python3.7/site-packages/torch/include/ATen/Tensor.h:12,^M from /home/guangrui/anaconda3/lib/python3.7/site-packages/torch/include/ATen/Context.h:4,^M from /home/guangrui/anaconda3/lib/python3.7/site-packages/torch/include/ATen/ATen.h:5,^M from /home/guangrui/anaconda3/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/types.h:3,^M from /home/guangrui/anaconda3/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/data/dataloader_options.h:4,^M from /home/guangrui/anaconda3/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/data/dataloader/base.h:3,^M from /home/guangrui/anaconda3/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/data/dataloader/stateful.h:3,^M from /home/guangrui/anaconda3/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/data/dataloader.h:3,^M from /home/guangrui/anaconda3/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/data.h:3,^M from /home/guangrui/anaconda3/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/all.h:4,^M from /home/guangrui/anaconda3/lib/python3.7/site-packages/torch/include/torch/extension.h:4,^M from csrc/flatten_unflatten.cpp:1:^M /home/guangrui/anaconda3/lib/python3.7/site-packages/torch/include/ATen/core/dispatch/Dispatcher.h: In instantiation of ‘Return c10::Dispatcher::doCallUnboxedOnly(const c10::DispatchTable&, const c10::LeftRight<ska::flat_hash_map<c10::TensorTypeId, c10::KernelFunction> >&, Args ...) const [with Return = void; Args = {const at::Tensor&, const at::Tensor&, bool, bool}]’:^M /home/guangrui/anaconda3/lib/python3.7/site-packages/torch/include/ATen/core/dispatch/Dispatcher.h:201:114: required from ‘Return c10::Dispatcher::callUnboxedOnly(const c10::OperatorHandle&, Args ...) const [with Return = void; Args = {const at::Tensor&, const at::Tensor&, bool, bool}]’^M /home/guangrui/anaconda3/lib/python3.7/site-packages/torch/include/ATen/core/TensorMethods.h:66:75: required from here^M /home/guangrui/anaconda3/lib/python3.7/site-packages/torch/include/ATen/core/dispatch/Dispatcher.h:211:80: error: redeclaration of ‘const at::Tensor& args#0’^M 211 | return kernel.template callUnboxedOnly<Return, Args...>(std::forward<Args>(args)...);^M | ^~~~^M /home/guangrui/anaconda3/lib/python3.7/site-packages/torch/include/ATen/core/dispatch/Dispatcher.h:209:115: note: ‘const at::Tensor& args#0’ previously declared here^M 209 | c10::optional<TensorTypeId> dispatchKey = dispatchTable.dispatchKeyExtractor().getDispatchKeyUnboxed<Args...>(args...);^M | ^~~~^M /home/guangrui/anaconda3/lib/python3.7/site-packages/torch/include/ATen/core/dispatch/Dispatcher.h:211:80: error: redeclaration of ‘const at::Tensor& args#1’^M 211 | return kernel.template callUnboxedOnly<Return, Args...>(std::forward<Args>(args)...);^M | ^~~~^M /home/guangrui/anaconda3/lib/python3.7/site-packages/torch/include/ATen/core/dispatch/Dispatcher.h:209:115: note: ‘const at::Tensor& args#1’ previously declared here^M 209 | c10::optional<TensorTypeId> dispatchKey = dispatchTable.dispatchKeyExtractor().getDispatchKeyUnboxed<Args...>(args...);^M | ^~~~^M /home/guangrui/anaconda3/lib/python3.7/site-packages/torch/include/ATen/core/dispatch/Dispatcher.h:211:80: error: redeclaration of ‘bool& args#2’^M 211 | return kernel.template callUnboxedOnly<Return, Args...>(std::forward<Args>(args)...);^M | ^~~~^M /home/guangrui/anaconda3/lib/python3.7/site-packages/torch/include/ATen/core/dispatch/Dispatcher.h:209:115: note: ‘bool& args#2’ previously declared here^M 209 | c10::optional<TensorTypeId> dispatchKey = dispatchTable.dispatchKeyExtractor().getDispatchKeyUnboxed<Args...>(args...);^M | ^~~~^M /home/guangrui/anaconda3/lib/python3.7/site-packages/torch/include/ATen/core/dispatch/Dispatcher.h:211:80: error: redeclaration of ‘bool& args#3’^M 211 | return kernel.template callUnboxedOnly<Return, Args...>(std::forward<Args>(args)...);^M | ^~~~^M /home/guangrui/anaconda3/lib/python3.7/site-packages/torch/include/ATen/core/dispatch/Dispatcher.h:209:115: note: ‘bool& args#3’ previously declared here^M 209 | c10::optional<TensorTypeId> dispatchKey = dispatchTable.dispatchKeyExtractor().getDispatchKeyUnboxed<Args...>(args...);^M

At the end:
` /home/guangrui/anaconda3/lib/python3.7/site-packages/torch/include/c10/util/LeftRight.h:67:10: error: ‘typename std::result_of<F(const T&)>::type c10::LeftRight::read(F&&) const [with F = c10::Dispatcher::doCallUnboxedOnly(const c10::DispatchTable&, const c10::LeftRight<ska::flat_hash_map<c10::TensorTypeId, c10::KernelFunction> >&, Args ...) const [with Return = at::Tensor; Args = {const at::Tensor&, c10::ArrayRef, c10::ArrayRef, c10::ArrayRef, c10::ArrayRef, c10::ArrayRef}]::<lambda(const ska::flat_hash_map<c10::TensorTypeId, c10::KernelFunction>&)>; T = ska::flat_hash_map<c10::TensorTypeId, c10::KernelFunction>; typename std::result_of<F(const T&)>::type = at::Tensor]’, declared using local type ‘c10::Dispatcher::doCallUnboxedOnly(const c10::DispatchTable&, const c10::LeftRight<ska::flat_hash_map<c10::TensorTypeId, c10::KernelFunction> >&, Args ...) const [with Return = at::Tensor; Args = {const at::Tensor&, c10::ArrayRef, c10::ArrayRef, c10::ArrayRef, c10::ArrayRef, c10::ArrayRef}]::<lambda(const ska::flat_hash_map<c10::TensorTypeId, c10::KernelFunction>&)>’, is used but never defined [-fpermissive]^M
/home/guangrui/anaconda3/lib/python3.7/site-packages/torch/include/c10/util/LeftRight.h:67:10: error: ‘typename std::result_of<F(const T&)>::type c10::LeftRight::read(F&&) const [with F = c10::Dispatcher::doCallUnboxedOnly(const c10::DispatchTable&, const c10::LeftRight<ska::flat_hash_map<c10::TensorTypeId, c10::KernelFunction> >&, Args ...) const [with Return = at::Tensor&; Args = {at::Tensor&, const at::Tensor&, c10::ArrayRef, c10::ArrayRef, c10::ArrayRef, c10::ArrayRef}]::<lambda(const ska::flat_hash_map<c10::TensorTypeId, c10::KernelFunction>&)>; T = ska::flat_hash_map<c10::TensorTypeId, c10::KernelFunction>; typename std::result_of<F(const T&)>::type = at::Tensor&]’, declared using local type ‘c10::Dispatcher::doCallUnboxedOnly(const c10::DispatchTable&, const c10::LeftRight<ska::flat_hash_map<c10::TensorTypeId, c10::KernelFunction> >&, Args ...) const [with Return = at::Tensor&; Args = {at::Tensor&, const at::Tensor&, c10::ArrayRef, c10::ArrayRef, c10::ArrayRef, c10::ArrayRef}]::<lambda(const ska::flat_hash_map<c10::TensorTypeId, c10::KernelFunction>&)>’, is used but never defined [-fpermissive]^M
/home/guangrui/anaconda3/lib/python3.7/site-packages/torch/include/c10/util/LeftRight.h:67:10: error: ‘typename std::result_of<F(const T&)>::type c10::LeftRight::read(F&&) const [with F = c10::Dispatcher::doCallUnboxedOnly(const c10::DispatchTable&, const c10::LeftRight<ska::flat_hash_map<c10::TensorTypeId, c10::KernelFunction> >&, Args ...) const [with Return = at::Tensor; Args = {const at::Tensor&, c10::ArrayRef, c10::ArrayRef, c10::ArrayRef, c10::ArrayRef}]::<lambda(const ska::flat_hash_map<c10::TensorTypeId, c10::KernelFunction>&)>; T = ska::flat_hash_map<c10::TensorTypeId, c10::KernelFunction>; typename std::result_of<F(const T&)>::type = at::Tensor]’, declared using local type ‘c10::Dispatcher::doCallUnboxedOnly(const c10::DispatchTable&, const c10::LeftRight<ska::flat_hash_map<c10::TensorTypeId, c10::KernelFunction> >&, Args ...) const [with Return = at::Tensor; Args = {const at::Tensor&, c10::ArrayRef, c10::ArrayRef, c10::ArrayRef, c10::ArrayRef}]::<lambda(const ska::flat_hash_map<c10::TensorTypeId, c10::KernelFunction>&)>’, is used but never defined [-fpermissive]^M
error: command 'gcc' failed with exit status 1^M
Running setup.py install for apex ... ^[[?25l^[[?25herror^M
Rolling back uninstall of apex^M
Moving to /home/guangrui/anaconda3/lib/python3.7/site-packages/apex-0.1-py3.7.egg^M
from /tmp/pip-uninstall-v72mrjmz/apex-0.1-py3.7.egg^M
Replacing /home/guangrui/anaconda3/lib/python3.7/site-packages/apex-0.1-py3.7.egg from /tmp/pip-uninstall-v72mrjmz/apex-0.1-py3.7.egg^M
Rolling /home/guangrui/anaconda3/lib/python3.7/site-packages/easy-install.pth back to previous state^M
Cleaning up...^M
Removing source in /tmp/pip-req-build-_dh7ztxj^M
Removed build tracker: '/tmp/pip-req-tracker-o_dho7t1'^M
^[[31mERROR: Command errored out with exit status 1: /home/guangrui/anaconda3/bin/python -u -c 'import sys, setuptools, tokenize; sys.argv[0] = '"'"'/tmp/pip-req-build-_dh7ztxj/setup.py'"'"'; file='"'"'/tmp/pip-req-build-_dh7ztxj/setup.py'"'"';f=getattr(tokenize, '"'"'open'"'"', open)(file);code=f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, file, '"'"'exec'"'"'))' --cpp_ext --cuda_ext install --record /tmp/pip-record-gcgz29ft/install-record.txt --single-version-externally-managed --compile --install-headers /home/guangrui/anaconda3/include/python3.7m/apex Check the logs for full command output.^[[0m^M
Exception information:^M
Traceback (most recent call last):^M
File "/home/guangrui/anaconda3/lib/python3.7/site-packages/pip/_internal/cli/base_command.py", line 186, in _main^M
status = self.run(options, args)^M
File "/home/guangrui/anaconda3/lib/python3.7/site-packages/pip/_internal/commands/install.py", line 404, in run^M
use_user_site=options.use_user_site,^M
File "/home/guangrui/anaconda3/lib/python3.7/site-packages/pip/_internal/req/init.py", line 71, in install_given_reqs^M
**kwargs^M
use_user_site=options.use_user_site,^M
File "/home/guangrui/anaconda3/lib/python3.7/site-packages/pip/_internal/req/init.py", line 71, in install_given_reqs^M
**kwargs^M
File "/home/guangrui/anaconda3/lib/python3.7/site-packages/pip/_internal/req/req_install.py", line 829, in install^M
scheme=scheme,^M
File "/home/guangrui/anaconda3/lib/python3.7/site-packages/pip/_internal/operations/install/legacy.py", line 72, in install^M
cwd=install_req.unpacked_source_directory,^M
File "/home/guangrui/anaconda3/lib/python3.7/site-packages/pip/_internal/utils/subprocess.py", line 275, in runner^M
spinner=spinner,^M
File "/home/guangrui/anaconda3/lib/python3.7/site-packages/pip/_internal/utils/subprocess.py", line 242, in call_subprocess^M
raise InstallationError(exc_msg)^M
pip._internal.exceptions.InstallationError: Command errored out with exit status 1: /home/guangrui/anaconda3/bin/python -u -c 'import sys, setuptools, tokenize; sys.argv[0] = '"'"'/tmp/pip-req-build-_dh7ztxj/setup.py'"'"'; file='"'"'/tmp/pip-req-build-_dh7ztxj/setup.py'"'"';f=getattr(tokenize, '"'"'open'"'"', open)(file);code=f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, file, '"'"'exec'"'"'))' --cpp_ext --cuda_ext install --record /tmp/pip-record-gcgz29ft/install-record.txt --single-version-externally-managed --compile --install-headers /home/guangrui/anaconda3/include/python3.7m/apex Check the logs for full command output.^M
1 location(s) to search for versions of pip:^M

• https://pypi.org/simple/pip/^M
  Fetching project page and analyzing links: https://pypi.org/simple/pip/^M
  Getting page https://pypi.org/simple/pip/^M
  Found index url https://pypi.org/simple^M
  Getting credentials from keyring for https://pypi.org/simple^M
  Getting credentials from keyring for pypi.org^M
  Starting new HTTPS connection (1): pypi.org:443^M
  https://pypi.org:443 "GET /simple/pip/ HTTP/1.1" 200 12868^M`

Start by reproducing the pip install command and inspect setup.py, csrc/flatten_unflatten.cpp, and the PyTorch headers named in the compiler output. Compare the build configuration for PyTorch 1.4.0 and CUDA 10.1; done means the Apex extensions compile and install without the reported errors.